Emergence of Chemotherapy Resistance in Cancer: Microenvironments, Genomics, and Game Theory Approaches

Abstract

Most cancers are still incurable because resistance to therapy is inevitable. Cancer cells usually acquire chemotherapy resistance due to two properties of cancer: adaptive cellular response to a heterogeneous microenvironment, and nonlinear interactions among various types of cells in a tumor community. In this thesis we construct in vitro heterogeneous tumor microenvironments to gain physiologically relevant information of phenotypic properties of cancer, cancer genomes, and interactions among various cells. Here we focus on metastatic breast cancer and multiple myeloma, a top five common cancer and blood cancer, respectively.

We first design drug gradient devices to mimic a tumor microecology during chemotherapeutic treatment, and assess multi-day spatio-temporal dynamics of breast cancer cells. Elevated resistance to doxorubicin (a chemotherapeutic drug) of breast cancer cells has been observed in a doxorubicin gradient based on proliferation rate, cell morphology, and cell motility. We test the hypothesis of horizontal gene transfer in breast cancer as a mechanism to diversify a population and enhance cellular adaptability to drug.

We then investigate genomic aspects of the rapid emergence of 16-fold elevated doxorubicin resistance in multiple myeloma (MM), which is achieved in a doxorubicin gradient within two weeks. We analyze RNA-sequencing data of the emerged resistant MM against non-resistant MM. Strikingly, we discover that mutational cold spots are ancient genes, maintaining the tness of cells and playing an important role in elevated drug resistance. Furthermore, we probe the interacting population dynamics of MM and bone marrow stromal cells in a doxorubicin gradient. By developing a spatial model inspired by game theory, we successfully predict the future densities of multiple myeloma and stromal cells in such heterogeneous environment.

Finally, we suggest that our approaches, including microfluidics experiments, next-

generation sequencing analyses, and quantitative modeling, can provide deeper insights on the emergence of therapy resistance in cancer and implications of novel therapy design.