Intermittent targeted therapies and stochastic evolution in patients affected by chronic myeloid leukemia

Abstract

Abstract. Front line therapy for the treatment of patients aected by chronic myeloid leukemia (CML) is based on the administration of tyrosine kinase inhibitors, namely imatinib or, more recently, axitinib. Although imatinib is highly eective and represents an example of a successful molecular targeted therapy, the appearance of resistance is observed in a proportion of patients, especially those in advanced stages. In this work, we investigate the appearance of resistance in patients aected by CML, by modeling the evolutionary dynamics of cancerous cell populations in a simulated patient treated by an intermittent targeted therapy. We simulate, with the Monte Carlo method, the stochastic evolution of initially healthy cells to leukemic clones, due to genetic mutations and changes in their reproductive behavior. We first present the model and its validation with experimental data by considering a continuous therapy. Then, we investigate how fluctuations in the number of leukemic cells aect patient response to the therapy when the drug is administered with an intermittent time scheduling. Here we show that an intermittent therapy (IT) represents a valid choice in patients with high risk of toxicity, despite an associated delay to the complete restoration of healthy cells. Moreover, a suitably tuned IT can reduce the probability of developing resistance.

Introduction: Unsolved Problems on Noise

“…Fluctuations in much more complex biological systems, like our brain, require however more phenomenological rules. A paradigmatic example of the usefulness of noise models in our own health is shown in [32] where the role of fluctuations on cancer growth dynamics is presented. As another example, the role that noise plays in the transmission of information in the brain is studied in [33].…”

The Mean First-Passage Time and Stochastic Resonance of a Stochastic Tumor-immune Model via BPNN Method

“…On the one hand, the theoretical methods, such as finite difference method [20,21], path integral method [22] and Galerkin method [23,24], have strict requirements for the dimension and complexity of the model and are not universally applicable. On the other hand, the simulation mainly refers to Monte-Carlo method [25], which is based on the Law of Large Numbers, requires a large number of sampling paths, so the accuracy is easily affected by the amount of data. Therefore, it is urgent to find a new algorithm to solve MFPT with strong applicability and robustness.…”

Stochastic and parameter analysis for an integrative cancer model

“…Developing a model for any biological system as a deterministic system, or at least not exposed to small random events, is something that can lead to a representation that is far from reality [7,8]. Usually, to model biological processes, such as a process of tumour development and cancer, stochasticity is used with the addition of different appropriate types of noise [9,10].…”