Applications de la modélisation mathématique à l'optimisation des traitements chimiothérapiques des gliomes de bas-grade

Pauline Mazzocco 1
1 NUMED - Numerical Medicine
UMPA-ENSL - Unité de Mathématiques Pures et Appliquées, Inria Grenoble - Rhône-Alpes
Abstract : Low-grade gliomas are slow-growing brain tumors, mainly affecting young adults who may remain without any symptoms for years. Patients can undergo surgery, or receive radiotherapy or chemotherapy with two different treatments: PCV of temozolomide (TMZ).In our different projects, we aim to show that mathematical modeling, and population approach, can allow to improve treatments, in terms of response duration and amplitude, for low-grade gliomas treated with chemotherapy (PCV and TMZ).In a first part, we focus on the possibility to modify PCV administration protocol, on a population level, to prolong tumor decrease duration. We claim that prolonging time interval between cycles enables us to significantly postpone the time to tumor regrowth.In a second part, we study the evolution of low-grade gliomas treated with TMZ. We analyze tumor size observations of 77 low-grade glioma patients, as well as genetic information, to develop a K-PD mixed-effects model describing tumor evolution before and after treatment onset. We then evaluate model capacity to predict tumor response duration and amplitude, on the base of early tumor sizes and genetic information. These predictions could be used to help clinicians to determine if they should prolong the treatment or not, for a given patient.In a last part, we more particularly focus on the phenomenon of resistance to TMZ. We build a PK-PD mixed-effects model describing the emergence of resistant tumor cells, using the same tumor size observations as previously. This model more accurately reproduces the evolution of TMZ in the body and its effect on the tumor. It is then used to optimize TMZ therapeutic protocol, on an individual level. Using an optimization algorithm, we determine the time interval between TMZ cycles, and the dose to administer, to prolong tumor decrease duration while limiting the emergence of resistance. The optimized protocols are evaluated with a stochastic approach, allowing to test the robustness of the model and the optimization.Through these different projects, we show the utility of mathematical modeling to help to improve chemotherapy treatments of low-grade glioma patients. We believe that these results could be transposed to other types of cancers.
Complete list of metadatas
Contributor : Abes Star <>
Submitted on : Monday, November 30, 2015 - 11:52:13 AM
Last modification on : Thursday, November 21, 2019 - 1:36:19 AM
Long-term archiving on: Tuesday, March 1, 2016 - 12:30:46 PM


Version validated by the jury (STAR)


  • HAL Id : tel-01235541, version 1



Pauline Mazzocco. Applications de la modélisation mathématique à l'optimisation des traitements chimiothérapiques des gliomes de bas-grade. Equations aux dérivées partielles [math.AP]. Université Grenoble Alpes, 2015. Français. ⟨NNT : 2015GREAM022⟩. ⟨tel-01235541⟩



Record views


Files downloads