Modelling Colorectal Cancer in the Mouse
Colorectal cancer (CRC) is the second leading cause of cancer-related death in the Western world, and despite a thorough understanding of the genetic events associated with the disease, we have very few effective non-surgical treatment options. A fundamental problem in identifying new targets for drug development is defining which, if any, of the many genetic and epigenetic changes seen in human tumors, represent a critical requirement for tumor cell survival and growth. To combat this problem, we develop unique genetically engineered mouse models (GEMMs) that accurately reflect the tumor-associated changes seen in human cancers, yet allow precise molecular manipulation of specific gene targets (Dow and Lowe, 2012). These models allow us to define how specific genetic defects influence disease initiation, progression and response to therapy, and ultimately, how we could more effectively treat malignant disease.
Defining genetic drivers of CRC
The genetic basis of cancer is extremely complex. Large-scale genomic analyses continue to describe hundreds of genomic alterations associated with cancer, including mutations, deletions, duplications, amplifications, translocations, and inversions. However, in most cases we have yet to clearly define those genetic (or epigenetic) events that represent critical cancer drivers, and equally importantly, those which are clinically actionable. We are addressing this challenge using tailored in vivo and ex vivo model systems to thoroughly characterize candidate driver events identified through next generation sequencing, and develop valid preclinical models to evaluate treatment strategies. To do this on a more rapid scale than previously possible, we have pioneered the use of shRNA (Premsrirut and Dow et al, 2011) and CRISPR/Cas9 technologies in mice (Dow and Fisher et al, 2015) to generate complex genetic models in a fraction of the time of traditional GEMMs. Most recently we developed a series of optimized Cas9 and base editing tools to enable simple genome modification in cells organoids, and mice (Zafra et al, 2018). Our tools enable the parallel investigation of single or multiple genes as well as complex chromosomal rearrangements (Han et al, 2017) that, until now, have been extremely difficult to model experimentally. Our ultimate goal is a thorough and detailed understanding of the genetic triggers of CRC that can inform treatment strategy.
Role of APC mutations and Wnt signaling in CRC progression and maintenance
APC mutation and/or elevated WNT signaling are observed in nearly all CRCs. We have developed a number of unique tools and animal models to explore the importance of Apc loss and Wnt activation in driving tumorigenesis, to address whether this signaling network represents a viable target for cancer therapy. Most recently, we have used a regulated shRNA approach (Premsrirut and Dow et al, 2011, Dow et al, 2012) to show that Apc loss is absolutely essential for the survival and growth of colorectal tumors, including those that harbor additional oncogenic insults such as Kras (Dow et al, submitted). Using our newly developed genome editing tools (Zafra et al, 2018), we are now focused on understanding how Wnt signaling thresholds and specific Apc mutations influence disease progression and response to therapy, to better define how Wnt-targeted drugs might be applied clinically.
Modeling targeted therapy in humans and mice
Forward genetic screens continue to identify many exciting new candidate drug targets for cancer treatment. Yet, in most cases, there are very few, if any, small molecule compounds for a given target that can be used to assess efficacy and toxicity in pre-clinical models. We take a genetic approach to explore target efficacy and tumor response, using shRNA-mediated gene silencing to mimic drug-induced target inhibition (Bolden, Tasdemir and Dow et al, 2014). The flexibility of shRNA-driven silencing provides the opportunity to assess the potential for ANY gene target and simultaneously characterize the effect of systemic gene inhibition – identifying potential toxicities associated with a given treatment. Combined with the analysis of primary human CRCs, we aim to define tumor vulnerabilities that can streamline the process of targeted drug development.