Recent evidence suggests that nongenetic (epigenetic) mechanisms play an important role at all stages of cancer evolution. In many cancers, these mechanisms have been observed to induce dynamic switching between two or more phenotypic states, which commonly show differential responses to drug treatments. Understanding how the rates of cell proliferation and phenotypic switching vary between individual cancers is crucial to understand how different cancers evolve and to inform individualized treatment strategies. In this talk, I will present a new statistical framework for estimating these parameters, using data from commonly performed cell line experiments, where phenotypes are sorted and expanded in culture. The framework explicitly models the stochastic dynamics of cell division, cell death and phenotypic switching, and it provides likelihood-based confidence intervals for the model parameters. The input data can be either the fraction of cells or the number of cells in each state at one or more time points. In the talk, I will emphasize discussing the identifiability of the model parameters, meaning which parameters can be estimated accurately depending on what data is collected. I will also discuss from an evolutionary perspective how knowledge of the model parameters can guide the selection of optimal treatment strategies, either involving an anti-cancer agent alone or a combination of an anti-cancer agent and an epigenetic drug. Finally, I will discuss the importance of rigorously quantifying the uncertainty in the estimation for understanding how treatment affects the evolutionary dynamics of the tumor and for assessing the robustness of any treatment recommendations.