Towards a quantitative understanding of tumor evolution

点击翻译按钮获取中文摘要

ABSTRACT Cancers are dynamic biological entities whose clonal architecture can change under strong selection pressures, such as exposure to therapy. As tumors progress through different stages, they coevolve with stromal populations, which in aggregate constitutes a significant challenge in assessing the potential value of new therapeutic strategies. Recent discoveries by us and other groups have provided a glimpse of the complexity of the clonal architecture of many tumors, their dynamics under therapy, and their interactions with the immune system. In most tumors, several sub-clonal populations simultaneously co-exist, and initially minor clones play a dominant role in subsequent phases of the tumor’s evolution. As clonal and stromal heterogeneity emerge as driving forces underlying cancer progression and therapeutic failure, there is a critical need for uncovering the quantitative fundamental principles underlying the evolution of tumors and their dynamic interaction with their microenvironment. My recent work has shown that tumor evolution does not proceed in a stochastic fashion but through a highly structured process, and that future dominant subclones can both be identified and targeted. The quantitative approaches developed by my group in the last few years are particularly tailored to elucidate the evolutionary patterns of clonal systems under strong selection. The central hypothesis of this proposal is that (1) tumor and stroma coevolve in an orchestrated fashion, (2) seeding clones can be identified through genomic and single cell longitudinal sampling, (3) these clones can be targeted, and (4) in order to characterize these clones we need to develop new quantitative approaches. The overarching goal of the present proposal is to uncover the mechanisms by which small tumor and stromal populations coevolve and drive tumor progression and the emergence of drug resistance, using glioblastoma as a paradigm. Quantitative approaches and fundamental principles of tumor evolution derived from this research will then be applied to other tumor types.