Ganesh Rao - US grants

DESCRIPTION (provided by applicant): Glioma, the most common primary brain tumor, remains incurable. Developing effective treatments depends on a better understanding of relevant cellular programs responsible for generating these tumors. An emerging concept in cancer research contends that activation of proliferative cellular processes is insufficient to cause tumor progression without concomitant suppression of apoptosis. The primary objective of this research proposal is to investigate the contribution of anti-apoptotic genes on the initiation, maintenance, and progression of glioma. Genes in the Signal Transducer and Activator of Transcription (STAT) signaling pathway are overexpressed in glioma and this pathway is a central hub of multiple cellular programs relevant to gliomagenesis including apoptotic suppression. The STAT3 gene, in particular, is associated with the aggressive mesenchymal subtype of glioma. Our hypothesis is that apoptotic suppression mediated by the STAT signaling pathway plays a causal role in the inexorable malignant progression of glioma. We will test this hypothesis by expressing genes in the STAT signaling axis in vivo to elucidate the effects of STAT pathway activation on glioma progression. Currently, most research evaluating gene overexpression in vivo is performed with xenograft models in immunodeficient mice that fail to recapitulate the microenvironment of brain tumors. These models use tumors formed by the implantation of fully malignant cells, rather than arising from a transformational event in a normal cell, thus obscuring analysis of the impact of a gene on the critical stages a tumor must overcome during its evolution. To study how anti-apoptotic genes affect tumor development we will employ a method of somatic cell transfer using the RCAS/tv-a system. This model permits the study of gene expression on endogenous tumor formation from its putative cell of origin in an immunocompetent mouse. Importantly, this model can be used to study the effect of immunosuppression on glioma progression as tumor-induced immunosuppression is mediated by STAT signaling. Our preliminary studies indicate that the anti-apoptotic genes in the STAT signaling pathway (including Bcl-2 and STAT3) enhance tumor formation, decrease survival, facilitate immunosuppressive intra-tumoral macrophages, and increase malignant progression by inducing necrosis - the hallmark of high-grade glioma. To investigate the impact of anti-apoptotic signaling programs and characterize their function, genes in the STAT signaling axis will be expressed using the RCAS/tv-a system. Specifically, we will investigate the roles of the Bcl2 (Specific Aim 1) and STAT gene families (Specific Aim 2) on tumor formation and progression. We expect to discover their contribution to the malignant degeneration of glioma and their promotion of tumor- induced immunosuppression. Ultimately, our results will help define therapeutic targets for glioma and the model will be used to test novel therapeutics against this deadly disease. PUBLIC HEALTH RELEVANCE: Understanding the relevant genetic aberrations that contribute to brain cancer is critical for developing new therapies for this deadly disease. The signal transducers and activators of transcription (STAT) pathway is a regulator of multiple cellular processes critical to brain tumor formation. We will study the effect of STAT signaling on brain tumor initiation and progression and test novel therapeutics against this pathway with a unique mouse model that recapitulates critical features of the human disease.

SUMMARY Glioma is the most common and deadliest primary brain tumor in humans. Highly malignant gliomas often arise from more indolent lower grade gliomas. Although patients with low-grade gliomas (LGG) may survive for many years, their tumors almost inevitably progress to high-grade gliomas (HGG), after which death occurs in 12 to 15 months. The process of malignant progression is poorly understood. Our published studies (funded by a Mentored Clinical Scientist Program [K08]) showed that anti-apoptotic signaling plays a key role in facilitating the progression of LGG to HGG. We also showed that suppression of apoptosis caused profound immunosuppression in the tumor microenvironment. Furthermore, we have shown, using several immunotherapeutic strategies, that reversing intratumoral immunosuppression can mitigate malignant progression in a murine model of glioma. We now hypothesize that antiapoptotic signaling promotes malignant progression in glioma by inducing an immunosuppressive tumor microenvironment. A major obstacle to studying malignant progression has been the lack of matched patient samples of LGGs and the HGGs to which they progress. However, we have identified over 250 patients who were treated for both LGG and later HGG at MD Anderson Cancer Center. The analysis of matched tumor samples from these patients represents a unique opportunity for the study of malignant progression. In the proposed work, we will take advantage of next- generation sequencing (NGS) to investigate the mechanisms through which LGG degenerates to HGG. In Aim 1, we will use NGS to identify anti-apoptotic genes that are overexpressed in HGGs relative to LGGs. A functional analysis of these genes in an immune competent murine model of glioma will determine their immunosuppression- and malignant transformation?promoting effects. In Aim 2, we will study two antiapoptotic genes (MCL-1 and BIRC3) that have emerged as lead facilitators of immunosuppression from analysis of TCGA LGG and HGG expression data as well as our own internal cohort of patients. We will model these genes in vivo to determine their impact on malignant progression. In Aim 3, we will profile our specimens to identify transcription factors that activate chemokines known to cause the intratumoral influx of key immunosuppressive cells. These transcription factors will be modeled in vivo to determine their effect on malignant progression. Identifying the factors that contribute to malignant progression will potentially enable us to mitigate the causes of progression. Thus, tumors may be maintained in the more indolent low-grade state rather than progressing to HGG, significantly prolonging survival. Ultimately, our results may also be applicable to other tumor types that demonstrate progression from a low- to high-grade lesion. This work is being done in collaboration with recognized experts in gene expression profiling, computational biology, biostatistics, and brain tumor immunology. We will also leverage MD Anderson?s Sequencing and Microarray Facility.

SUMMARY Glioma is the most common and deadliest primary brain tumor in humans. Highly malignant gliomas often arise from more indolent lower grade gliomas. Although patients with low-grade gliomas (LGG) may survive for many years, their tumors almost inevitably progress to high-grade gliomas (HGG), after which death occurs in 12 to 15 months. The process of malignant progression is poorly understood. Our published studies (funded by a Mentored Clinical Scientist Program [K08]) showed that anti-apoptotic signaling plays a key role in facilitating the progression of LGG to HGG. We also showed that suppression of apoptosis caused profound immunosuppression in the tumor microenvironment. Furthermore, we have shown, using several immunotherapeutic strategies, that reversing intratumoral immunosuppression can mitigate malignant progression in a murine model of glioma. We now hypothesize that antiapoptotic signaling promotes malignant progression in glioma by inducing an immunosuppressive tumor microenvironment. A major obstacle to studying malignant progression has been the lack of matched patient samples of LGGs and the HGGs to which they progress. However, we have identified over 250 patients who were treated for both LGG and later HGG at MD Anderson Cancer Center. The analysis of matched tumor samples from these patients represents a unique opportunity for the study of malignant progression. In the proposed work, we will take advantage of next- generation sequencing (NGS) to investigate the mechanisms through which LGG degenerates to HGG. In Aim 1, we will use NGS to identify anti-apoptotic genes that are overexpressed in HGGs relative to LGGs. A functional analysis of these genes in an immune competent murine model of glioma will determine their immunosuppression- and malignant transformation?promoting effects. In Aim 2, we will study two antiapoptotic genes (MCL-1 and BIRC3) that have emerged as lead facilitators of immunosuppression from analysis of TCGA LGG and HGG expression data as well as our own internal cohort of patients. We will model these genes in vivo to determine their impact on malignant progression. In Aim 3, we will profile our specimens to identify transcription factors that activate chemokines known to cause the intratumoral influx of key immunosuppressive cells. These transcription factors will be modeled in vivo to determine their effect on malignant progression. Identifying the factors that contribute to malignant progression will potentially enable us to mitigate the causes of progression. Thus, tumors may be maintained in the more indolent low-grade state rather than progressing to HGG, significantly prolonging survival. Ultimately, our results may also be applicable to other tumor types that demonstrate progression from a low- to high-grade lesion. This work is being done in collaboration with recognized experts in gene expression profiling, computational biology, biostatistics, and brain tumor immunology. We will also leverage MD Anderson?s Sequencing and Microarray Facility.

SUMMARY: ANIMAL CORE (CORE D) Investigations into brain tumor biology and therapy are best carried out in orthotopic animal models that mimic the natural milieu of the tumor. The ability to reach meaningful conclusions from these in vivo studies is greatly enhanced when uniform, readily reproducible animal models are used. The core also provides support for investigators using genetically engineered mouse models (in particular the RCAS/Ntv-a model). The purpose of the Animal Core is to provide an animal modeling service that centralizes expertise thereby achieving uniformity and reproducibility that permits accurate comparisons between experiments, research groups, and projects in this Brain Cancer SPORE. In the current funding period, the Animal Core performed 546 experiments involving over 15,000 animals for SPORE investigators. In the next 5-year funding period, the Animal Core will continue to provide expert services to all investigators in Brain Cancer SPORE. To this end, the Core will: 1) Provide support for orthotopic animal experiments using patient-derived glioma stem cell lines (GSCs), which are currently considered the gold standard in vivo model of human gliomas. This Core has developed approximately 100 GSCs, over 40 of which have been fully molecularly characterized; 2) Provide support for flank and orthotopic animal experiments using patient-derived tumor explants from the operating room implanted into NOD.Cg-PrkdcscidIl2rgtm1Wjl/SzJ (NSG) mice; 3) Provide support for animal experiments using immunocompetent genetically engineered mouse models (GEMMs), particularly the RCAS/Ntv-a system; 4) Provide support for orthotopic animal experiments testing oncolytic adenoviruses using immunocompetent hamster models that are permissive to viral replication; 5) Provide support for orthotopic animal experiments using traditional profession glioma cell lines from humans (U87, U251, LN229) and mice (GL261); 6) Provide support for in vivo imaging either with bioluminescence imaging (BLI) or magnetic resonance imaging (MRI) to assess efficacy of therapeutic strategies. These services are mission critical and one or more of these services are used by all Projects in this SPORE.

Project Summary Very limited options are available for treating glioblastoma, the most common primary brain tumor in humans. Effective surgical options are particularly lacking, although resection has been shown to consistently be of value for some patients with glioma. Laser interstitial thermal therapy (LITT) is in clinical use for treating primary brain tumors, but how this technology affects the tumor microenvironment is poorly understood. We have generated an immunocompetent RCAS/Ntv-a murine model of LITT with survivable brain lesions that can be used to characterize LITT-induced changes in the tumor microenvironment. Importantly, we have extensive experience studying the tumor microenvironment in the context of endogenously forming, high-grade gliomas in this mouse model. We hypothesize that LITT-induced thermal damage can create a tumor microenvironment more responsive to adjunct therapies. In Specific Aim 1, we will characterize the longitudinal effects of LITT on the tumor microenvironment by examining treated mice for an influx of immune cells and induced genetic changes using NanoString technology. We will also use a murine anti-PD-1 antibody, which we have recently shown to be effective against glioblastoma in our tumor model, in neoadjuvant and adjuvant settings to determine if its efficacy can be enhanced by LITT. While anti-PD-1 monotherapy for glioblastoma has not been efficacious due to the low immunogenicity of the tumor environment, its use in the context of LITT-induced immune cell infiltration and neoantigen formation may lead to greater therapeutic benefits against this type of cancer. In Specific Aim 2, we will determine the ability of thermally-released doxorubicin from nanoparticles to improve survival rates of tumor-bearing mice following LITT. Although in clinical trials for extracranial cancers, the use of heat-activated nanoparticles for treating brain tumors is quite novel. Systemic doxorubicin has shown some benefit in other murine models of brain cancer, but its heat-activated nanoparticle release may permit more localized delivery and extended treatment beyond the LITT penumbra to the infiltrating edge of the tumor, which is the most common source of glioblastoma recurrence. With the completion of these aims, we will better understand how the population immune cells in the tumor microenvironment changes in response to thermal therapy. We will also understand what genetic programs are upregulated in the tumor microenvironment after thermal therapy potentially giving us new therapeutic targets to combine with LITT. The overall goal of this proposal is to demonstrate how thermal ablation affects the tumor microenvironment and how it can be combined with other treatments to improve outcomes for patients with glioblastoma. Given the availability of the treatments being investigated there is a low threshold for the clinical application of our results. These studies will serve as the groundwork for more extensive studies on the use of LITT for the treatment of brain tumors.

ABSTRACT Immunotherapy has revolutionized cancer treatment by reversing the immune suppression of cytotoxic anti- tumor CD8+ effector T cells. However, glioblastoma patients have profound lymphopenia and immune checkpoint inhibition treatment does not restore T cell immune function. Extensive characterization by our group has demonstrated that glioblastoma is fundamentally different relative to other malignancies in its preferential enrichment of innate immune cells such as macrophages and microglia that are recruited to the tumor microenvironment. These innate immune cells are tumor supportive. In a genetically-engineered mouse model (GEMM) of glioblastoma we recapitulated lympophenia using a CD8 knockout (KO) background, and found marked enrichment of PD-1 expressing macrophages in the murine glioma microenvironment ? similar to observations made in human glioblastoma patients. We evaluated the effect of anti-PD-1 Ab delivered intravenously in glioblastoma-bearing wild-type mice and in the CD8 KO background and found therapeutic benefit even in the absence of the CD8 effector T cell. Both peripheral monocyte-derived macrophages and resident microglia were reduced within the glioblastoma microenvironment in mice treated with the anti-PD-1 Ab. As such, our overall study hypothesis is that anti-PD-1 exerts therapeutic immune modulatory effects against glioblastoma through innate immunity in the central nervous system (CNS). This proposal will address multiple crucial questions to the field including: 1) Does the anti-PD-1 Ab cross into the CNS to exert a therapeutic effect; 2) what immune cells, other than the CD8 T cell, are contributing to the therapeutic effect of this agent; 3) are the immune cells that are mediating the therapeutic effect arising from the periphery or are they intrinsic to the CNS; and 4) how does the glioblastoma immune microenvironment change in response to treatment? To address these questions, we will use contemporary murine models of glioma that closely approximate human glioblastoma and manipulate both the innate and adaptive immune systems to dissect the impact and importance of each in the context of anti-PD-1 treatment. Validation will be carried out using data from human subjects treated with anti-PD-1. By clarifying the mechanistic role of anti-PD-1 therapeutic activity, we may identify the subset of glioblastoma patients that are capable of responding to this type of strategy. This is a significant area of unmet need if glioblastoma patients are to benefit from immunotherapy. These studies may also reveal that anti-PD-1 treatment has a dual role on both the innate and adaptive immune system and when one arm is not operational this agent toggles its modulatory properties to the dominant immune arm.