Toward simple "in silico" experiments for drugs administration in some cancer treatments

TL;DR

This paper introduces a novel in silico framework combining flatness-based and model-free control to optimize cancer drug treatments, ensuring safety and adaptability despite uncertainties, demonstrated through numerical simulations.

Contribution

It presents a new combined control framework for designing cancer treatment schedules that handle uncertainties and reduce drug injections.

Findings

Health indicators reach safe levels in simulations.

The approach adapts to uncertainties in drug doses.

Some cases require no chemotherapeutic agents.

Abstract

We present some "in silico" experiments to design combined chemo- and immunotherapy treatment schedules. We introduce a new framework by combining flatness-based control, which is a model-based setting, along with model-free control. The flatness property of the used mathematical model yields straightforward reference trajectories. They provide us with the nominal open-loop control inputs. Closing the loop via model-free control allows to deal with the uncertainties on the injected drug doses. Several numerical simulations illustrating different case studies are displayed. We show in particular that the considered health indicators are driven to the safe region, even for critical initial conditions. Furthermore, in some specific cases there is no need to inject chemotherapeutic agents.