Abstract LB-011: Experimentally derived multi-state model of chemoresistance to characterize phenotypic dynamics in breast cancer

Abstract

Abstract Although several molecular mechanisms of chemotherapeutic resistance are well studied, the ability to model and predict the dynamic development of resistance in cancer remains elusive. The development of resistance to chemotherapy is a major cause of treatment failure in breast cancer. Understanding of the dynamic progression of resistance in vitro is significant both in optimizing clinical dosing schedules and in probing the fundamental rules governing cancer cell population dynamics. While mathematical models to predict the dynamics of heterogeneous cancer cell populations have been proposed, none have been experimentally tested or linked to an observable resistant phenotype. There exists a need to integrate predictive modeling with measurable phenotypic time-resolved cell population characteristics. In this work, we describe and model in vitro phenotypically heterogeneous cancer cell populations cultured in a homogeneous microenvironment. MCF-7 cells were treated with a 24-hour pulse of doxorubicin and monitored over a fifteen-week period. Samples were counted each week, drug sensitivity was assayed with an AOPI live-dead reagent, and DNA content was measured. Phase images taken each week were used to classify cells according to morphology and determine subpopulation frequency. Drug sensitivity data were used to calibrate three mathematical models to interrogate the models’ ability to capture time-resolved response to drug. While all three models capture a trajectory in which population level resistance increases following drug exposure and eventually attenuates, results indicated that a multi-state model which incorporates the role of heterogeneity and cellular plasticity is superior at capturing the phenotypically observed response. Here we describe a two state model that provides an experimentally derived estimate of how cancer cells populate resistant and sensitive states over time following a drug perturbation. Drug sensitivity, proliferation rate, DNA content, and morphology frequency shift were quantified with proliferation rate dropping, the proportion of cells with 4N or higher DNA content increasing, and the proportion of cells displaying typical cuboidal morphology decreasing following doxorubicin treatment. Within approximately 6-7 weeks, proliferation rates, DNA content, and morphology distributions return to baseline, consistent with observed alterations in drug sensitivity. In future work, this approach could be extended to other cancer cell lines and eventually have a direct clinical application. This contribution is the first work to our knowledge that combines experimental time-resolved drug sensitivity data with a mathematical model of resistance development. This work aims to provide a platform for future investigation into the dynamic role of phenotypic heterogeneity in the development of chemotherapeutic resistance in all aspects of cancer research and therapy. Citation Format: Kaitlyn E. Johnson, Grant Howard, Amy Brock, Thomas Yankeelov. Experimentally derived multi-state model of chemoresistance to characterize phenotypic dynamics in breast cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr LB-011. doi:10.1158/1538-7445.AM2017-LB-011