Oncogenic mechanisms, molecular stratification and therapeutic targets of brain tumors

点击翻译按钮获取中文摘要

Abstract The work to be pursued in this application will continue and expand the program pioneered by Dr. Iavarone to combine innovative computational tools and state-of-the-art experimental cancer models in vitro and in vivo to identify homogeneous subgroups of cancer patients in order to dissect the pathogenesis of cancer and design tailored and fully validated personalized therapeutic approaches. The application is focused on glioblastoma multiforme, one of the most lethal forms of human cancer. The investigation of glioblastoma has represented a long-standing effort of Dr. Iavarone’s laboratory, which in recent work has produced novel targeted therapeutic opportunities currently being tested in clinical studies. The proposal will also benefit from the organizational contexts recently set in motion by the large network operations coordinated by the PI. The research plan is articulated around the development of a novel and integrated computational-experimental framework for: i) the identification of homogeneous groups of tumors sharing activation of the same biological pathways; ii) the study of cancer heterogeneity at the single cell level to accurately inform tumor classifications; iii) the therapeutic prediction emerging from the identification of driver modules and synthetic lethal relationships of malignant glioma. We will develop and apply novel technologies for high-throughput transcriptomic and proteomic analysis of individual cells within malignant glioma tissues. These approaches, which we have pioneered in our laboratory at Columbia University during the last few years, will serve as the basis for the multifaceted computational analysis that will extract genes and proteins responsible for the phenotypic state of individual cells. Experimental validations will be selectively applied to the novel and most exciting molecular pathways and will be performed by our laboratory that has an array of experimental tools and sequence-annotated patient-derived models to pursue each individual question. As for the selection of oncogene-dependent and independent vulnerabilities identified by our previous work, the ability of our studies to identify novel driver phenotypes and master regulators of individual tumor cells will be geared towards routing the new mechanisms into pathway-based synthetic lethality that will inform specific drug sensitivities. The successful outcome of this proposal is an integrated computational-experimental pipeline that will be able to mechanistically identify the determinants of tumor genomes and phenotypes of solid tumors. This information will be of invaluable significance to decipher evolving tumor dependencies and provide the most accurate therapeutic predictions.