DESARROLLO E IMPLEMENTACIÓN DE UN SISTEMA DE TOMA DE DECISIONES PARA LA OPTIMIZACIÓN DE TRATAMIENTOS QUIMIOTERAPÉUTICOS CONTRA EL CÁNCER: SoT-Q

Abstract

The aim of this work is to develop a decision support system (SoT-Q) to assist medical oncologists in the optimization of the dosing schedules of the cytotoxic agents used in the chemotherapy treatments administered to their patients. To accomplish this, two modules developed in C++ were integrated: a mathematical modeling module, that couples the Gompertz growth model for tumor masses with the pharmacokinetic model of the administered drugs; and, a second module, which allows multi-objective optimization, through the use of the genetic algorithm SPEA2, where, the established optimization objectives were: (1) to minimize the size of the tumor (number of cells present in the tumor mass); (2) to minimize the toxicity of the treatment based on the cumulative dose of each drug at the end of the treatment, and (3) to minimize its total cost. To test the system, the treatments generated by SoT-Q were compared with the corresponding treatments proposed in the Manual of Clinical Oncology from Brazil (MOC), for three different types of cancer: lung cancer (NSCLC), breast cancer (HER-2) and head and neck cancer (HNSCC); finally, significant improvements regarding to the proposed optimization objectives were achieved.