Bone metastasis treatment modeling via optimal control (Q1738035)

This paper deals with two optimal control models related to two treatments for bone metastasis: denosumab and radiotherapy. Under some assumptions, the authors prove the existence and uniqueness of solutions, $C(t),$ $B(t)$, $T(t)$, representing the population of the osteoclasts (OCs), of the osteoblasts (OBs) and the of bone metastatic cells (CCs), respectively, as functions of time in $]0,t_f[$, to the minimization of appropriate cost functional, reflecting the economical cost and the side-effects, subject to the ordinary differential equations: \[\dot{C}=\alpha_1(1-u_D)CB^{g_1}-(\beta_1+\psi_1u_R) C+\sigma_1CT, \] \[\dot{B}=\alpha_2C^{g_2}B-(\beta_2+\psi_2u_R) B+\sigma_2BT,\] \[\dot{T}=\alpha_3T(1-T/m)-(\beta_3+u_R) T+\sigma_3C^{g_2}T +\sigma_4CT^{g_1}.\] Here, $u_D$ represents the effect of denosumab on the activity of the OCs and $u_R$ represents the cell-killing rate due to the radiation on the basic multicellular unit in the radiotherapy treatment model. The constants $\alpha_i$ and $\beta_i$ are the rates of cell production and removal, respectively, for $i=1,2,3$, $g_1<0$ and $g_2>0$ are the net effectiveness of paracrine factors, $\sigma_i$ ($i=1, 2, 3, 4$) are the proportional rates of the OCs-CCs and OBs-CCs interactions, $\psi_1$ and $\psi_2$ are radiotherapy control parameters, and $m$ is the carrying capacity of the logistic growth rate of CCs within the BMU location. The study of optimal control solutions is given by using the Pontryagin's Maximum Principle. Numerical results are provided for three scenarios: (1) Agressive metastasis, $\alpha_3=1.5\times 10^{-2}$, $\sigma_1=1.0\times 10^{-6}$, $\sigma_3=1.0\times 10^{-3}$ and $\sigma_4=0.0$; (2) OCs-dependent metastasis, $\alpha_3=1.0\times 10^{-4}$, $\sigma_1=1.0\times 10^{-6}$, $\sigma_3=1.0\times 10^{-3}$ and $\sigma_4=0.0$; (3) Slow, BMU-dependent metastasis, $\alpha_3=1.0\times 10^{-4}$, $\sigma_1=0.0$, $\sigma_3=1.0\times 10^{-8}$ and $\sigma_4=- 1.0\times 10^{-4}$.