1. Examining mechanisms of therapeutic response and resistance

Overview: We are exploring the idea that cancers can resist therapeutic treatment by rapidly adopting new cell states that differ in their therapeutic sensitivity. Toward this goal, we use state-of-the-art experimental approaches to systematically examine the cell states operable in triple-negative breast cancers, identify the relationships between cell states and the tumor microenvironment, and design therapeutic strategies to target cell state vulnerabilities and promote anti-tumor microenvironments. This project is informed by the Knight Cancer Institute SMMART clinical trials program, which is testing rapid, personalized therapies that are watched closely via recurring biopsies and deep analysis.  

Partners:
Lisa M. Coussens, M.D. (h.c.), Ph.D., FAACR
Gordon Mills, M.D.,Ph.D.

Related papers:
1. "Multiplex spatial systems analysis of local nanodose drug responses predicts effective treatment combinations of immunotherapies and targeted agents in mammary carcinoma," Nature Biotechnology
2. "A Novel Mouse Model that Recapitulates the Heterogeneity of Human Triple Negative Breast Cancer," Nature Communications

2. Modeling tumor ecosystem responses

Overview: Recent studies have demonstrated that some therapies targeting intrinsic programs in malignant epithelial cells also induce broad changes to multiple aspects of the tumor ecosystem and that treatment efficacy can be improved by selecting drug combinations that simultaneously attack malignant cells in vulnerable states, disrupt pro-tumor microenvironments, and create sustained anti-tumor microenvironments. We are addressing this problem by developing agent-based models (ABMs) comprised of key cell types in tumor ecosystems, how they interact, and the consequences of these interactions on therapeutic response. Our ultimate goal is to enable computational selection of treatments designed to maximize tumor control by directly targeting malignant cells and by promoting anti-tumorigenic microenvironments. ABMs model individual cells as independent “agents” that can interact, adapt to defined perturbations, and migrate through a spatial model following a set of biologically-motivated “rules” that define their actions and interactions with each other and signals in their environment. This computational framework provides a quantitative time- and cost-effective approach to study tumor ecosystem interactions and to identify experimentally-testable hypotheses. 

Partners:
Young Hwan Chang, Ph.D.
Paul Macklin, Ph.D.

Related papers:
1. "Digitize your Biology! Modeling multicellular systems through interpretable cell behavior," bioRxiv preprint

3. Understanding the role of cellular heterogeneity in cancer

Overview: Cancers display substantial cell-to-cell variability, but little is known about how this variation influences therapeutic response. Enabled by our live-cell imaging approaches, we have found that organizing individual cells into lineage families (parents, siblings, cousins, etc.) provides a robust framework to examine cellular heterogeneity. We are developing novel computational models to identify drug combinations predicted to target distinct cell lineage states, which we test experimentally. This project is designed to uncover the molecular underpinnings of heritable heterogeneity, which will reveal insights into mechanisms that drive key cancer hallmarks. Ultimately, our findings could be used to develop new therapies designed to favorably control adaptive lineage responses. 

Partner:
Aaron Meyer, Ph.D.

Related papers:
1. "Analysis and modeling of cancer drug responses using cell cycle phase-specific rate effects," Nature Communications
2. "A lineage tree-based hidden Markov model to quantify cellular heterogeneity and plasticity," Communications Biology