Optimal Intervention in Markovian Gene Regulatory Networks With Random-Length Therapeutic Response to Antitumor Drug

Abstract—The most effective cancer treatments are the ones that prolong patients ’ lives while offering a reasonable quality of life during and after treatment. The treatments must also carry out their actions rapidly and with high efficiency such that a very large percentage of tumor cells die or shift into a state where they stop proliferating. Due to biological and microenvironmental variabili-ties within tumor cells, the action period of an administered drug can vary among a population of patients. In this paper, based on a recently proposed model for tumor growth inhibition, we first probabilistically characterize the variability of the length of drug action. Then, we present a methodology to devise optimal inter-vention strategies for any Markovian genetic regulatory network governing the tumor when the antitumor drug has a random-length duration of action. Index Terms—Cancer therapy, gene regulatory networks (GRNs), optimal intervention, probabilistic Boolean networks (PBNs), tumor growth inhibition (TGI) model. I