Phenotypic Profiling of Colon Tumor Organoid Monolayers Characterizes a FOLFOX-Induced, AKT-Driven Subpopulation of Chemoresistant Cells

Abstract

Colorectal cancer (CRC) is the third most commonly diagnosed cancer in the US as well as the third most common cause of cancer-related death. Standard treatment includes surgical resection and adjuvant chemotherapy with multidrug regimens such as FOLFOX (Folinic Acid, 5-Fluorouracil, Oxaliplatin). However, due to high heterogeneity in colon tumors, patient response is variable and tools to predict drug response are few. Additionally, resistance, which can lead to recurrence and metastasis, is not uncommon. As such, there is a need for improved treatment options, better methods for predicting drug response, and better understanding of resistance mechanisms in CRC. Here, we introduce a host of new methodologies to phenotypically characterize CRC with the long term goals of better understanding chemotherapy resistance and creating actionable clinical tools that can model and predict individual tumors’ drug response. We have curated a set of colon tumor organoids, titled the Heterogeneity Library, that represents a variety of CRC subtypes. We used our newly designed phenotypic profiling pipeline, which leverages DAPI, pAKT, CD44v9, and H2AX, to characterize a FOLFOX-induced, PI3K/AKT-driven survival signature originally identified in our transcriptomic data. Unbiased subpopulation analysis confirms a cohort of cells with high PI3K/AKT signaling, high DNA damage, and high levels of stem cell markers across organoids. We believe that cells increasingly upregulate survival signaling, including extracellular matrix components and growth factor receptors, which all activate AKT, which in turn leads to a protective, resistant phenotype. Ultimately, we posit the observed FOLFOX dose-dependent increases in the fraction of these resistant cells across tumor types is a combination of adaptation down this proposed Resistance Trajectory in combination with a selection process whereby susceptible cells are readily eliminated, leaving the increasingly mesenchymal progenitor- and stem-like cells to expand despite FOX exposure. Finally, we endeavored to target this resistant population by treating organoid monolayers with a combination of FOLFOX and Dactolisib, the PI3K/mTOR dual inhibitor, which exhibited significant synergy across organoids. Even the most FOLFOX-resistant organoid – made up almost entirely of the previously described highly damaged, AKT-driven, stem-like resistant cells – demonstrated increased FOLFOX efficacy when also treated with Dactolisib. We hope that with further validation, the novel contributions of this dissertation can be developed into actionable tools for use from bench to bedside.