Abstract 2690: Qualitatively distinct evolutionary games played by EGFR driven NSCLC

Background: Assuming cell-autonomous fitness within a population is often insufficient for understanding phenotypic shifts and resulting equilibria that can occur within ecosystems. Non-cell-autonomous interactions, whether competitive or cooperative in nature, influence the evolutionary trajectory of the population as a whole. We represented the interactions between gefitinib resistant and sensitive lung adenocarcinoma cells as evolutionary games to better understand population shifts that occur within heterogeneous tumors across a range of microenvironmental contexts. Methods: Co-cultures of fluorescently labeled PC-9 parental and evolved resistant lineages were plated across a range of relative initial proportions in 384 and 96 well formats. Bright field, GFP, and Texas Red images of each well were obtained in 3-4 hour intervals using an automated cell culture system. Least squares regressions were performed on the growth rate against the initial proportion of subpopulations in each well. Evolutionary games were quantified by converting the linear equations into payoff matrices correspond to each of the different conditions. Results: Interactions between the parental and resistant types within 11 different environmental drug conditions (DMSO, gefitinib, etoposide, erlotinib, doxorubicin, paclitaxel, gemcitabine, olaparib, osimertinib, lapatinib, and neratinib) were characterized and the corresponding payoff matrices derived. These games were plotted in a game space that lets us easily distinguish qualitatively distinct games. The distribution of each condition within the space was derived from the uncertainty of the preceding analysis. For example, while the interaction profile within gefitinib completely localizes to the “Deadlock” region and olaparib to the “Battle of the Sexes” region, the range of games played within paclitaxel crosses the boundary between the “Hawk-Dove” and “Concord” regions. Discussion: Our measurement of growth trajectories of the two different types across the different environmental contexts provided insight into the range of intercellular interactions that can occur in the presence of the various selective pressures. The qualitatively distinct regions of the constructed game space were predictive of the resulting phenotypic equilibrium of the heterogeneous co-cultures. For example, games played within the “Deadlock” region invariably lead to extinction of the parental lineage; however, interactions within the “Battle of the Sexes” region allowed for quantitatively predictable coexistence of both subpopulations. As such, the relative positions of the different eco-evolutionary dynamics correspond to different drugs within this space can be utilized to develop therapeutic strategies that steer the interactions (i.e. ‘treat the game9) into regions that are associated with more therapeutically favorable equilibrium. Citation Format: Nathan Farrokhian, Artem Kaznatcheev, Patrick Ellsworth, Masahiro Hitomi, Emily Dolson, Jacob G. Scott. Qualitatively distinct evolutionary games played by EGFR driven NSCLC [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 2690.