2018 Research Project Summaries

RESEARCH COLLABORATION GRANT

Anders Persson, PhD and Andrew Chi, MD, PhD

Synthetic lethal targeting of NAD(P)H-dependent DNA damage in IDH mutant gliomas
With ~20,000 new cases diagnosed in the United States each year, gliomas are the most common malignant primary brain tumor in adults. About a third of gliomas have mutations in the isocitrate dehydrogenase (IDH) genes, specifically IDH1 or IDH2. Although patients with IDH mutations have a better prognosis, nearly all IDH mutant gliomas regrow after radiation and chemotherapy. If the tumor regrows, no therapy is effective and most patients die from their tumor. The lack of relevant IDH mutant glioma models available to study has been a roadblock in developing new therapies for IDH mutant gliomas. However, the Persson and Chi laboratories have successfully established {tooltip}cell cultures{end-texte}Artificial system to grow and study cells in a controlled environment, outside of a human or animal.{end-tooltip} and animal models using patient-derived IDH mutant gliomas. Recent studies suggest that current therapies that specifically target the IDH1 mutation may not be an effective way to control gliomas. Therefore, our laboratories have used an alternative strategy to identify vulnerabilities of IDH mutant glioma cells..
  • Beta-lapachone: A naturally occurring compound obtained from the bark of the lapacho tree that has cancer preventive properties and is currently tested in clinical trials
  • NAD+: An important molecule involved in metabolism
The Persson laboratory found that beta-lapachone reduced levels of NAD+ and weakened the function of IDH1 mutant glioma cells. The Chi laboratory discovered that IDH mutant gliomas are extremely vulnerable to NAD+ reduction. To further our studies, we will:
  1. Determine whether processing of beta-lapachone by the enzyme NQO1 induce cell death of IDH1 mutant glioma cells and whether NQO1 is a biomarker for response to beta-lapachone treatment
  2. Identify other agents that reduce NAD+ and other drugs that work cooperatively with beta-lapachone to specifically kill IDH mutant gliomas
We hope to validate beta-lapachone and drug combinations as candidate therapies for IDH mutant glioma patients in desperate need of effective treatment.

BASIC RESEARCH FELLOWSHIPS

Zhenyi An, PhD

Targeting TLR2 in EGFR/EGFRvIII+ glioblastoma
Glioblastoma (GBM) is the most common malignant brain tumor in adults and there is currently no cure. Increased copies of the genes named EGFR and EGFRvIII are commonly seen in GBM tumors.
  • About a third of GBM patients who have extra copies of EGFR also have extra copies of EGFRvIII.
  • Almost all GBM tumors with EGFRvIII have extra copies of EGFR, indicating that these two molecules might work together to promote tumor growth.
Our preliminary data showed that in GBM cells that have EGFR and EGFRvIII, the protein TLR2 is critical for EGFR and EGFRvIII to send signals inside the cells. We also found that TLR2 can change what types of {tooltip}immune cells{end-texte}Cells found in the body that fight off infections and disease.{end-tooltip} are found in tumors, causing an increase the number of immune cells called “macrophages.” Macrophages help tumors grow and act as obstacles for immune-therapy. In this study, we will:
  1. Investigate the role TLR2 has in regulating tumor growth and the types of immune cells in the tumor (using GBM patient samples, GBM cell lines and well-established GBM mouse models).
  2. Test if blocking TLR2 would be an effective treatment for GBMs that have increased copies of EGFR and EGFRvIII, and if blocking TLR2 can be effectively combined with immune-therapy.
Successful completion of this research will shed light on new therapeutic strategies for GBMs with EGFR and EGFRvIII.

Wei Du, MD, PhD

Unraveling regulatory T cell-dependent anti-tumor mechanisms in brain metastasis
The metastasis (or spreading) of cancer cells to the brain is a threatening reality for cancer patients that can result in a dismal outcome and severe neurological symptoms. Unfortunately, current therapies for brain metastases offer only minimal benefits. In order to develop new therapies for prevention and treatment, we need to further understand of how these cancer cells set up tumors in the brain. Regulatory T cells (Treg): A type of {tooltip}immune cell{end-texte}Cells found in the body that fight off infections and disease.{end-tooltip} that blocks other immune cells and can help tumor growth. In primary and metastatic tumors, a high level of Treg cells correlates with poor prognosis. While the brain has many cell types that affect metastatic tumor growth, we hypothesize that Treg cells promote brain metastasis by modifying the brain environment in a way that makes it easier for tumors to grow. This research will provide much needed knowledge about the interactions between the immune system and the cancer cells that enter the brain and form tumors. A better understanding of these interactions will lead to new therapies that are greatly needed for metastatic breast cancer and melanoma patients.

Morgan Schrock, DVM, PhD

A novel antimitotic in glioblastoma
Only 3 percent of glioblastoma (GBM) patients live longer than five years, which stresses the desperate need for new therapeutic strategies.
  • Mitosis: the process of one cell becoming two
  • Antimitotics: drugs that block mitosis
Because mitosis is required for tumor growth, antimitotics are an effective treatment for many cancers because they prevent cancer cells from dividing into more cells. However, of the antimitotics currently available, none are suitable for use in the brain due to the development of neurologic side effects. I have identified a {tooltip}regulator of mitosis{end-texte}A molecule in cells that can speed up or slow down the process of cell division or even stop it completely.{end-tooltip}, mitotic kinesin-like protein 2 (MKlp2), that has yet to be greatly studied. Blocking MKlp2 significantly reduces the growth of GBM cells without the side effects that current antimitotics cause. These effects on GBM cells cannot be explained by the known function of MKlp2 during mitosis. Therefore, I am predicting a new function for MKlp2 in mitosis. Within this project, I propose to:
  1. Test this new critical role for MKlp2 in mitosis
  2. Determine the therapeutic potential of blocking MKlp2, in combination with standard GBM therapies
  3. Characterize MKlp2 expression in naturally-occurring canine cancers to determine whether treating canine brain tumors can serve as a preclinical study to testing MKlp2 inhibitors in GBM.
These studies will improve our understanding of GBM growth and the role of MKlp2 in GBM cells, in addition to providing the foundation for pushing a new antimitotic toward clinical trials.

Euhnee Yi, PhD

Tracing extrachromosomal DNA inheritance patterns in glioblastoma using CRISPR
Glioblastoma (GBM) is the most aggressive primary brain tumor; it has poor prognosis and frequently {tooltip}recurs{end-texte}Comes back.{end-tooltip} after therapy.
  • Tumor heterogeneity: Substantial differences in the cells that make up a tumor, like differences in the amount types of genetic material found in different cells within a single tumor.
Tumor heterogeneity is believed to be the reason why treatments for GBM fail, why tumors become resistant to therapy, and why cancers recur.
  • Extrachromosomal DNA (ecDNA): DNA that is not part of the chromosomes
Our previous study revealed that many cancer-causing genes are found in ecDNA. When tumor cells divide, ecDNA is not equally distributed between {tooltip}daughter cells{end-texte}The cells that are the result of one cell dividing into two cells.{end-tooltip} in the same way as traditional DNA; one daughter cell may get ecDNA while another does not. We hypothesize that this inconsistent pattern of ecDNA distribution contributes to tumor heterogeneity. Therefore, a better understanding of how ecDNA behaves in tumor cells may improve therapy development for GBM and other cancers. To address this, I will develop a new genetic tool that allows us to see the ecDNA in live GBM cells in order to track ecDNA functions over time. Results of this study will provide new insights into how ecDNA causes changes in tumors.

Fan Zhang, PhD

Programming tumor-clearing macrophages with targeted in situ gene therapy
Macrophages (mφs) are {tooltip}immune cells{end-texte}Cells found in the body that fight off infections and disease.{end-tooltip} that infiltrate into gliomas in high numbers. Upon reaching the tumor, they undergo a switch from an activated state, known as M1, to an {tooltip}immunosuppressive {end-texte}Blocks the normal disease-fighting functions of immune cells.{end-tooltip} state, called M2.
  • M1: Activated state of mφs that attack cancer cells
  • M2: Immunosuppressive state of mφs that helps tumors to grow into other parts of the brain
Current immunotherapies that target mφs block mφs in the whole body and cause dangerous side effects. Instead of total blockade of all mφs in the whole body, we propose to locally reprogram the M2 mφs in the tumor into the highly effective, tumor-clearing M1 mφs by using nanoparticles (NPs). The nanoparticles will deliver genes directly to the mφs, causing them to switch to the M1 state. By reprogramming mφs, the tumors may be more responsive to the currently available immunotherapies. This approach relieves glioma patients from wide-spread toxicities by only causing inflammation at the tumor site. The results of this study will:
  1. Provide evidence for using gene-modification systems to “correct” immune function within the tumors without resorting to treatments that broadly disrupt the immune system.
  2. Provide a basis for combining materials science, gene therapy, and immunology to develop new immunotherapies for the cancer treatment.

DISCOVERY GRANTS

Loic Deleyrolle, PhD

Targeting glioma slow-cycling cells using autologous dendritic cell vaccine
In the last 20 years, available therapies have done little to improve prognosis for glioblastoma patients. It has been proposed that the most important clinical target to improve disease outcome may be a specific fraction of tumor cells known as cancer stem cells.
  • Slow-cycling cells: Cells in a tumor that can go dormant, are resistant to traditional therapies, and have characteristics of cancer stem cells.
We previously demonstrated the presence of slow-cycling cells in glioblastoma, which serves as an enriched reservoir of cancer-promoting cells able to initiate new tumors and cause relapse of the disease. Therapies that target this specific cell type have the potential to reduce disease progression and recurrence. Moreover, using the immune system to identify tumor cells while preserving the surrounding normal tissue offers a clear benefit over using conventional treatments that typically are not able to target the tumor without affecting the surrounding healthy tissue. This project proposes to develop a new kind of immunotherapy, known as an autologous dendritic cell vaccine, that will use RNA from slow-cycling cells to activate immune cells. The goal of the proposal is to generate a strong immune response against the cancer-promoting slow-cycling cells. This work offers a shift in thinking about immune targeting of cancer and points out areas where new interventions may prove valuable.

Christian Grommes, MD

Characterizing and modulating the immune evasion landscape in CNS Lymphoma
Primary Central Nervous System Lymphoma (PCNSL) is a rare, but highly aggressive form of non-Hodgkin lymphoma that is only found in the central nervous system. Currently, treatment options are limited to chemotherapy and radiation. New potential treatment approaches use agents – called immune checkpoint inhibitors – that utilize the patient’s own immune system to battle cancer cells. Cancer cells can “hide” from {tooltip}immune cells{end-texte}Cells found in the body that fight off infections and disease.{end-tooltip} by displaying immune checkpoint proteins. By using immune checkpoint inhibitors to block these immune checkpoint proteins, cancer cells become “visible” to the immune system and can be killed. This concept has been very successful in treating Hodgkin lymphoma; however, whether or not PCNSLs use immune checkpoint proteins to go undetected from immune cells is largely unknown. The goal of this project is to learn more about the distribution of checkpoint markers in a large set of banked PCNSL tissues. To help conserve the precious tissue samples, we will use an innovative staining system that allows us to examine multiple markers in a single sample. By studying the tissues, we will:
  1. Be able to correlate the checkpoint protein analysis with DNA sequencing data that we already have for these tissues
  2. Test whether checkpoint inhibitors alter the levels of the checkpoint proteins using new models of PCNSL.
The ultimate aim is to help guide the selection of ideal immune checkpoint inhibitor(s) for future clinical trials in PCNSL patients.

Christopher Hubert, PhD

WDR5 is a Niche-Specific Vulnerability in Glioblastoma
The cells that make up glioblastoma (GBM) – the most common and severe type of brain cancer – have a significant genetic diversity, which is part of the reason GBM treatment options remain ineffective. This diversity is due to the way that tumor {tooltip}stem cells{end-texte}Stem cells are unspecialized cells that can divide to create new copies of themselves while also creating other more specialized cell types.{end-tooltip} can give rise to other tumor cell types, as well as the great variation in the conditions around individual GBM cells.
  • Glioma stem cells (GSCs): Tumor cellsthat have the ability to give rise to all cell types found in a particular tumor. The characteristics of GSCs can significantly vary between patients, and a single tumor may have several different types of GSCs.
  • Glioma stem cell ‘niche’: A distinct region and environment within a tumor where glioma stem cells are located. The niche helps in the maintenance of stem cell state and also plays a role in making the tumor resistant to therapies. There can be different kinds of niches harboring different types of GSCs in a single tumor.
Cells in different niches behave and respond differently to treatments. Finding new weaknesses in each separate population is critically important in order to develop more effective therapies. Recently, {tooltip}3D cell culture{end-texte}An artificially created environment in which cells are permitted to grow or interact with their surroundings.{end-tooltip} systems have enabled the growth and isolation of glioma stem cells within a system that mimics the GSC niches. This allows for separate glioma stem cell populations to finally be studied accurately. We have used these systems to pinpoint a particular protein (called WDR5) as key factor for GBM survival in our 3D models of niches that promote rapid tumor growth. In our studies, we plan to:
  1. Explore how WDR5 drives GBM growth
  2. Test a specific inhibitor of WDR5 in patient-derived pre-clinical models
Completing this proposal will lay a foundation for future studies into an unexplored area of GBM biology. At the same time, we will use highly accurate patient-derived pre-clinical models to test a potentially powerful therapeutic target that has never been considered in GBM.

Matthew Sarkisian, PhD

Primary Cilia-Derived ARL13B As A Biomarker and Promoter of Gliomagenesis
Gliomas are the most common form of brain cancer and can often be deadly. This proposal will examine the impact that the protein ARL13b has on glioma growth. While little is known about ARL13b, it is known to enhance the signaling of another protein called smoothened (SMO) by preventing its breakdown and keeping it active. Patients with gliomas that have high levels of ARL13b and SMO tend to have significantly shorter survival outcomes, suggesting the interaction between these two molecules makes tumors more aggressive.
  • Primary cilia: Part of the glioma cell that stems out from the cell surface, like an antennae or whiskers. Primary cilia control {tooltip}signaling pathways{end-texte}A group of molecules in a cell that work together to control cell function, such as cell division or cell death.{end-tooltip} that may regulate tumor cell reproduction, migration, survival, or other cell functions.
We found that increasing the levels of ARL13b in glioma cells causes an abnormal recruitment of SMO into the primary cilia. When found in primary cilia, SMO signaling is activated to increase the rate of tumor formation. Thus, when glioma cells have high ARL13b levels, they potentially hijack the SMO signaling in the cilia to amplify tumor growth. Further, when we live image tumor cilia that have excess ARL13b, we found for that cilia pinch off and release tiny packets containing ARL13b into their surrounding environment – a phenomenon that we believe encourages continued tumor cell growth. Thus, we hypothesize that, by promoting SMO signaling, ARL13b is a measurable {tooltip}biomarker{end-texte}An indicator (usually related to genetic information) that can be used to predict prognosis, response to treatment, or other information about a tumor{end-tooltip} and drives glioma growth. Our experiments will determine:
  1. If ARL13b levels outside of the cells increase as tumors grow
  2. If disabling ARL13b inhibits glioma growth or if it potentially enhances receptiveness to SMO inhibitors (which are already in development for cancer patients).

Stephanie Seidlits, PhD

A biomaterial approach to identify mechanopathology driving glioblastoma invasion
Glioblastoma (GBM) is a highly lethal brain cancer due to its aggressively invasive nature, multi-drug resistance and inevitable recurrence.
  • Glioma stem cells (GSCs): A subset of GBM cells that are resistant to treatment, highly invasive and thought to cause recurrence.
  • Extracellular matrix (ECM): The proteins and sugars that make up the space surrounding GSCs. The ECM assists in the migration of GSCs away from the primary tumor to other areas, where they then give rise to recurrent tumors.
In this project, we will use a {tooltip}cell culture{end-texte}An artificially created environment in which cells are permitted to grow or interact with their surroundings.{end-tooltip} platform of specially engineered biomaterials that mimic ECM components and vary between brain tumors and surrounding tissue. These biomaterial platforms will enable systematic characterization of:
  1. How the brain matrix drives GSC invasion
  2. How migrating GSCs alter their surroundings to promote tumor recurrence
  3. How treatment resistance affects these behaviors.
This biomaterial-based approach is less costly, more time efficient and better controlled than animal studies. Additionally, unlike other cell culture methods, this approach yields results with comparable clinical relevance. Ultimately, this research seeks to identify specific GSC-ECM interactions that lead to aggressive behaviors in tumors, and if disrupting these interactions could benefit patients by preventing or delaying tumor recurrence.

MEDICAL STUDENT SUMMER FELLOWSHIPS

Neil Almeida

Development of mass cytometry probes to assess function and phenotype of T-cells
{tooltip}Genetic variation{end-texte}Differences between cells in gene mutations or in genes that are turned on or off.{end-tooltip} among glioma cells within individual tumors is a major challenge. Immunotherapies a single antigen (a molecule that can activate an immune response) are likely to fail due to the growth of tumor cells that do not have that antigen as a result of genetic variation. In an attempt to overcome this challenge, the Okada lab at the University of California, San Francisco is currently conducting a clinical trial for low grade gliomas that using an immunotherapy directed at 10 antigens from a variety of proteins that are abundant in glioma cells. To monitor how each patient’s immune system recognizes each of the antigens, I used a cutting-edge technology called CyTOF mass cytometry that can detect and measure many different characteristics of cells at the same time. I have already developed probes called tetramers that detect {tooltip}immune responses{end-texte}When immune cells are activated to eliminate a specific target{end-tooltip} against the antigens. For this project, my goal is to apply this technology to analyze tumor and blood samples from the clinical trial patients in order to uncover the number and functionality of their immune cells. My objective will be to, for the first time, establish this technology as a method to simultaneously evaluate immune cell characteristics and the precision of the antigens.

Joshua Bernstock, PhD

Immune checkpoint blockade in combination with oHSVs for pediatric brain tumors
Primary malignant central nervous system tumors are the leading cause of cancer-related death and disability in children. While advances in surgery, radiation and chemotherapy have improved the survival rates of children with malignant brain tumors, extreme mortality persists in certain {tooltip}subpopulations{end-texte}Groups of patients whose tumors have similarities.{end-tooltip} and current therapies are often associated with extreme toxicity and life-long side effects. Genetically engineered oncolytic herpes simplex viruses (oHSVs) -a modified form of the viruses that produces common cold sores- are capable of selectively targeting and killing cancer cells and, therefore, have emerged as a promising therapeutic option for pediatric patient populations. This project will attempt to identify ideal pediatric patient groups for viral and immunotherapies, and in so doing explore the roll of {tooltip}immune checkpoint inhibitors{end-texte}A type of immunotherapy that blocks the molecules that cancer cells use to hide from the immune system.{end-tooltip} in combination with oHSVs (G207 and M032), which are currently being studied in clinical trials for some types of childhood and adult brain tumors. If successful, this research will provide evidence for rapid transition into clinical trials for children with high-grade brain tumors.

Saksham Gupta

Defining the Immunophenotype of Meningioma
Meningiomas are the most common primary brain tumor and can typically be completely treated with surgery. However, a small percentage of meningiomas recur and cannot be controlled with any available chemotherapies.
  • Immunotherapy: Treatments that use a person’s own immune system to target and fight cancer.
Immunotherapy has shown promise in treating many cancer types, but not yet in meningioma. This may be due to our poor understanding of what cells and molecules related to the immune system are found in these tumors. We will utilize a new technology to detect over 40 specific immune cell markers from a series of meningiomas that we are {tooltip}resecting{end-texte}Removing by surgery.{end-tooltip} from patients. In doing so, we will also determine whether these immune markers relate to the characteristics that are clinically assessed for each tumor including genetics and microscopic features This project will determine which of these components show promise for the development of new immune therapies that effectively target and kill meningioma cells.

Casey Jarvis

Effects of Host Pericyte Deficiency on Angiogenesis in Glioblastoma
Glioblastoma (GBM) is the most common and deadly type of primary brain cancer, with 12,500 new cases diagnosed each year. Even with aggressive therapy, most patients do not survive more than a year after diagnosis. Our lab is interested in studying the factors that make glioblastoma so malignant.
  • Angiogenesis: The formation of new blood vessels. Angiogenesis is essential for tumor growth.
  • Pericytes: {tooltip}Contractile{end-texte}Capable of contracting or causing contraction.{end-tooltip} cells that wrap around the cells that form our blood vessels to control the expanding and contracting of capillaries.
We hypothesize that pericytes play a role in initiating angiogenesis in GBM, and that a lack of pericytes will slow the formation of new blood vessels, thus limiting tumor growth. To test our hypothesis, we will implant GBM cells into mice that lack pericytes. We will then compare the degree of angiogenesis and tumor growth in these mice to tumors implanted in mice with normal pericyte levels. The data we collect is a preliminary study that will later be validated in a study using GBM samples received from patients. If we can show that a lack of pericytes does, in fact, block tumor growth, it would serve as evidence that new therapies that specifically target pericytes need to be developed. These therapies could have the potential to increase patient survival and reduce mortality from glioblastoma.

Adam Lauko

Effects of a Glioblastoma Secreted Cytokine on Myeloid-Derived Suppressor Cells
Glioblastoma (GBM) is the most common malignant brain tumor and continues to have poor prognosis, with survival averaging between 12 and 15 months. While advancements in chemotherapy and radiation have proven to be effective in other cancers, the use of these treatments has done little to increase glioblastoma survival rates. One explanation is that glioblastoma does an excellent job at shutting down the immune system at the tumor site, allowing the tumor to hide. Normally, after fighting an infection, cells known as myeloid-derived suppressor cells (MDSCs) deploy a full arsenal of proteins that shut down the immune system, preventing damage to healthy cells. We recently discovered that glioblastoma has figured out a way to hack this regulation, causing MDSCs to shut down the immune system at the tumor site, thus allowing the tumor to hide.
  • Macrophage migration inhibitory factor (MIF): The molecule that glioblastomas uses to activate MDSCs to shut down the immune system at the tumor site.
My project attempts to identify the way MIF activates MDSCs by accomplishing the following:
  1. Look at the signaling that occurs inside of MDSCs after MIF stimulates the cells.
  2. Determine which proteins in the MDSC arsenal are activated after MIF stimulation.
Both of these goals have the potential to identify future drug targets that would prevent the shutdown of the immune system. We believe this project will bring us closer to identifying novel treatments that use a person’s own immune system to target and fight cancer.

Gina Rhee

The contribution of stromal cell senescence to sex differences in glioblastoma
Men have a greater chance of developing brain tumors and a lower chance of survival compared to women. Additionally, the risk of brain tumors increases with age more rapidly for men than women. These patterns show that patient sex has a large influence on brain tumor formation and growth, particularly in age-associated processes. Over time, cell damage builds up in older cells, giving them the potential to transform into cancer cells. As a defense mechanism, our body “retires” these damaged cells, permanently halting their growth and division. With age, our body will gather more of these retired, or senescent, cells. Despite retirement, these cells still impact tumor formation and growth. Senescent cells release compounds that can either encourage or prevent the growth of nearby cells. In the brain, senescent cells may release different distinct compounds in men and women, which could contribute to men having a higher risk for brain tumors. While we know that senescent cells are an integral part of tumor formation, their exact role in the brain remains largely unknown. Thus, understanding this process is important as it may provide insight into new avenues for treatment. In this project, we will:
  1. Compare different properties of this retirement process in male and female brain cells using mouse models to determine if female brain cells undergo this process more easily than male brain cells.
  2. Examine brain tumor growth in the presence of male senescent cells, female senescent cells, and normal brain cells.

Yekaterina Shpanskaya

Radiomic Discovery of Novel Prognostic Imaging Biomarkers in Medulloblastoma
Medulloblastoma is the most common malignant brain tumor in children, and patient prognosis tends to vary. Currently, efforts for creating personalized medulloblastoma prognostic tools focus on molecular markers and can be costly, invasive, and require complex technologies that are not readily available to most hospitals. Magnetic Resonance Imaging (MRI) is conducted in all medulloblastoma patients as a primary method for tumor diagnosis and surveillance. We propose to identify accurate quantitative MRI markers that are predictive of medulloblastoma prognosis. To do this, we will use advanced computer-based imaging techniques combined with machine-learning algorithms. We have assembled medulloblastoma patient clinical, imaging, and survival data from five institutions in the United States and Canada. Our model will be trained and tested across this large patient population to assure accuracy, reliability and generalizability. When successful, our MRI prognostic markers have the potential to provide a clinically-applicable test that will be available to all patients at initial diagnosis and throughout clinical follow-up at no additional cost.

Johnathan Zeng

Optimizing a Radiomic Risk Score to Characterize Brain Tumor Progression in MRI
Radiation therapy is a common treatment that can be effective for patients with metastatic brain cancer. Unfortunately, radiation necrosis, or dead tissue, is a common side effect of radiation therapy. When viewing an MRI of the brain, tumors (both new and recurrent) and dead tissue look very similar, so it is difficult to distinguish one from another. Many times, more invasive methods, such as surgery, are used to confirm the diagnosis because the diagnosis will determine the next steps in treating the patient. Therefore, there is a need for an effective, less invasive method to distinguish new tumors from dead tissue. Our lab proposes to use a software called CoLIAGe, which uses advanced mathematics and machine learning to conduct a special imaging analysis on MRI scans to accurately predict whether or not a patient has a new or recurrent brain tumor. The goal of this project is to modify our CoLlAGe software so that it can be used on patient records in the hospital.