Tumors are complex systems composed of genetically and transcriptionally heterogeneous cells, and this variation has been implicated as a cause of drug resistance and mortality. Understanding intra-tumor heterogeneity is therefore likely to have widespread scientific implications and clinical applications. The advent of single-cell RNA-Seq has enabled the mapping of transcriptionally distinct states among cancer cells across a wide range of tumor types and disease stages. In this project, I take a gene module-centric view to define cell states in a rigorous and widely applicable manner. In my preliminary work, I have identified specific cancer cell states that are conserved across many cancer types as well as some that are unique to specific tumors. This proposal aims to further characterize these states in relation to the tumor microenvironment, with the goal of furthering our understanding of tumors as complex dynamical systems amenable to therapeutic intervention. In order to systematically map interactions between tumor cell populations within their native context, I will integrate paired single-cell and spatial transcriptomic data obtained in primary patient samples. This will enable me to identify cell populations that interact within the tumor, and to characterize the gene expression changes that occur in these interactions. In order to further establish how cancer cell state function relates to the tumor microenvironment, I will take advantage of experimental model systems amenable to perturbation. I will use orthotopic mouse cancer models as a platform to deplete specific populations of the immune system and measure their effect on cancer cell states. Next, I will perform co-culture experiments followed by transcriptomic and phenotypic profiling to measure direct effects on one cell population on another. Collectively, the experiments and algorithms proposed will significantly improve our understanding of intratumoral heterogeneity through the lens of cancer cell states. Dissecting the complex interactions between cancer cell states and immune cell populations will be of particular translational relevance, as it may enable the rational design of immunotherapy regimens. The research proposed capitalizes on the strengths of the Yanai lab in cutting-edge molecular techniques, computational innovation, and in vitro and in vivo modeling of cancer. Together with the outlined training plan, this work will set me on a path to independent research as a physician-scientist.
In this proposal, we investigate tumorigenesis through the lens of cancer cell states and their interaction with immune cells of the microenvironment. The proposal involves an interdisciplinary approach, integrating of today?s most cutting-edge experimental and computational approaches, including single-cell and spatial transcriptomics applied to patient samples and experimental models of tumorigenesis amenable to perturbation.
