Computational Analysis of Treatment Resistant Cancer Cells

ABSTRACT

Prostate cancer (PC), which is a disease driven by the activity of the androgen receptor (AR), is the most commonly diagnosed malignancy and despite advances in diagnostic and treatment strategies, PC is the second most common cause of cancer mortality in men (Bray et al., 2018). Taxane-based chemotherapy is the only chemotherapy that prolongs survival in metastatic PC patients (Petrylak et al., 2004; Tannock et al., 2004). At the cellular level, taxanes bind to and stabilize microtubules (MTs) inhibiting all MT-dependent intracellular pathways. MTs are highly dynamic polymers that stochastically switch between phases of growth, shrinkage, and pause (Jordan and Wilson, 2004). Altered MT dynamics endow cancer cells with both survival and migratory advantages (Mitchison, 2012). Taxanes inhibit MT dynamics and alter the spatial organization of the MT network, thereby inhibiting intracellular trafficking of molecular cargo critical for tumor survival. In PC specifically, taxanes inhibit transcriptional activity downstream of MT stabilization (Thadani-Mulero et al., 2012) and AR nuclear accumulation (Darshan et al., 2011; Zhu et al., 2010).

Different tubulin inhibitors, even from within the same structural class as the taxanes, affect distinct parameters of MT dynamics (Jordan and Wilson, 2004), yet the selection of taxane for chemotherapy is not based on the particular patterns of dynamic behavior of the MT cytoskeleton in individual patients. We envisage that systematic characterization using quantitative analysis of MT dynamics in PC patient cells expressing clinically relevant protein isoforms (Matov et al., 2024; Thoma et al., 2010), before and after treatment with each of the taxanes, will allow us to identify criteria for the selection of the most suitable drug combination at the onset of treatment. We link MT dynamics in the presence of AR variants and sensitivity/resistance to taxanes and connect fundamental research with clinically relevant concepts to elucidate cellular mechanisms of clinical response to taxanes and, thus, advance the customization of therapy. Our computational live-cell analysis addresses questions in the context of the inherent differences in MT homeostasis as a function of AR content in PC cells, the specific parameters of MT dynamics each of the taxanes affects, and how can this information be used to match endogenous patterns of MT dynamics with drug-modulated MT behavior. We investigate whether the sensitivity to taxanes, evaluated by computational analysis of MTs, can be linked to gene expression correlated with AR and its variants, and whether the resistance to taxanes can be linked to the presence of a specific AR splice variant, and can we identify which of the taxanes will be most effective based on the endogenous patterns of MT dynamics.