Thèse Molecular Design And Computational Strategies For Advanced Osteosarcoma Therapy Using Biomimetic Apatites And Drug Delivery Mosaic H/F Doctorat.Gouv.Fr

Établissement : Institut National Polytechnique de Toulouse École doctorale : SDM - SCIENCES DE LA MATIERE - Toulouse Laboratoire de recherche : CIRIMAT - Centre Interuniversitaire de Recherche et d'Ingénierie des Matériaux Direction de la thèse : Corinne LACAZE-DUFAURE ORCID 0000000331512685 Début de la thèse : 2027-09-01 Date limite de candidature : 2026-11-16T23:59:59 Osteosarcomas are bone tumours, primarily affecting children and adolescents, displaying a high risk of metastasis. Moreover, the 5-year survival rate is less than 25%, and high recurrence rates are observed. The clinical treatment combining surgery and chemotherapy faces two major problems: high doses of anticancer agents lead to severe side effects, which may be delayed, and the resistance of tumour cells to treatments is increasing.

Through specific computational studies, i.e. by using molecular dynamics and quantum calculations in the framework of DFT, we propose to investigate two alternatives to these problems: (i) a new route of treatment administration with vectorization using biomimetic synthetic apatites for local delivery in the bone of active molecules already used for osteosarcomas treatments, and (ii) a way to overcome resistance to anticancer agents by chemically modifying drug molecules and increasing their effectiveness. Overall, this will reduce the patient's drug doses. Thanks to computational simulations, biomimetic apatites for vectorization of anticancer agents or their model molecules will be studied in relation to delivery properties. In parallel, the mechanism of action (MoA) of these molecules will be studied at the molecular level to propose and validate modified anticancer agents. In a last step, classical molecular dynamics calculations will produce a new model of biomimetic apatites that can be used for the adsorption of large molecules. These computational studies will also provide answers regarding the possibility of co-adsorbing anticancer agents and other drug molecules (analgesics, painkillers, antibiotics...).

Developing computational models for the in-silico investigation of drug treatments is essential for accelerating and optimizing the development of new therapeutic strategies. In the specific context of osteosarcoma, we propose to investigate both optimal drug-delivery strategies and the design of chemically modified molecules. These computational studies will enable the identification and prioritization of the most promising candidates, thereby reducing the number of compounds for subsequent in vitro and in vivo tests.
Each year, about 400,000 children and adolescents aged 0-19 years old in the world develop cancer [1]. Osteosarcomas, which are bone tumours, account for approximately 20,000 cases per year, with 900 cases per year in Western Europe (including the EU, UK, Switzerland, and Norway) [2;3]. 70% of osteosarcomas occur in young people aged 10-25 years. For patients with such cancer still localized, the 5-year survival rate after simple tumour removal was previously 20%. It is currently around 70% at 5 years after a combination of surgery and chemotherapy or radiotherapy [4]. There is a high risk of metastasis, particularly to the lungs, which thus leads to a survival rate of less than 25% at 5 years [5]. Moreover, among patients who initially respond positively to treatment, recurrence is noted in almost 1 case out of 3 [6]. It is therefore vital to eliminate all cancerous cells: surgical removal of the bone tumour is followed by treatment with high doses of anticancer agents (methotrexate, doxorubicin, cisplatin...) [7]. These chemotherapy treatments cause severe side effects for the patient (nausea, heart problems, fatigue...). We propose to investigate solutions to these problems by specific computational studies.

The objectives of this PhD thesis are to:

(i) Boost targeted therapies by studying a new vector for anticancer molecules to reduce the administered doses and thus side effects: the mode of adsorption of anticancer molecules on biomimetic apatites will be determined at the molecular level with numerical quantum tools (Density Functional Theory, DFT). Indeed, bone tissues are constituted of an organic phase (collagen) with the insertion of apatite nanocrystals (mineral phase). Apatites consist of a sub-stoichiometric cristalline hydroxyapatite core (Ca2+, (PO4)3-, OH- ions), and an amorphous hydrated ionic layer (Ca2+, (HPO4)2- ions and water molecules) on the surface (Fig.1). Biomimetic synthetic apatites constitute thus non-toxic and effective bone substitutes. Their surface, which consists of a hydrated ionic layer, gives them high reactivity, necessary for the adsorption of anticancer agents and controlled delivery.
In previous projects, experimental studies conducted by the French team PPB of this project and Italian collaborators (M. Iafisco and A. Tampieri, ISTEC Faenza, and E. Boanini, Univ. Bologna) have already shown possible adsorption of methotrexate [8] and doxorubicin [9] on phospho-calcic apatites. The computational studies developed in this PhD thesis will determine the mode of adsorption of anticancer molecules on the biomimetic apatites that drives the delivery of drugs and anticancer agents. These computational studies will also address the possibility of co-adsorbing anticancer agents and other drug molecules (analgesics, painkillers, antibiotics...).

ii) Understand the mechanism of action of known anti-osteosarcoma agents and identify structural modifications that could lead to an increase in their efficacy and a reduction of their side effects.
Detailed study of drug mechanisms of action (MoA) helps establish clear structure-cytotoxicity relationships, enabling modification of known anticancer agents to increase efficacy and selectivity while reducing their well-known serious side effects. The standard treatments for curing osteosarcoma mainly use doxorubicin, cisplatin, and high-dose methotrexate, a standard approach that has not changed in more than 40 years [10]. In this PhD thesis, computational studies will provide a detailed understanding of the MoA of such drugs and allow the introduction of structural modifications to reduce the impact of side effects and to improve their efficacy.
The objectives of the PhD thesis are to:
* Boost targeted therapies by studying a new vector for anticancer molecules to reduce the administered doses and thus side effects: the mode of adsorption of anticancer molecules on biomimetic apatites will be determined at the molecular level with numerical quantum tools (Density Functional Theory, DFT).

* Understand the mechanism of action of known anti-osteosarcoma agents and identify structural modifications that could lead to an increase in their efficacy and a reduction of their side effects.
Methodology and expected results:

(i) Biomimetic apatite-based drug delivery: Quantum calculations + Classical Molecular Dynamics (MD) Simulation
A new model has recently been developed thanks to quantum calculations (Density Functional Theory DFT and ab initio MD, AIMD) at the molecular scale [11,12] (PhD thesis by Y. Hajji, supervisor Corinne Lacaze-Dufaure, PPB team of CIRIMAT Lab.). This model will initially be used to study the adsorption of molecules themselves (cisplatin) or smaller model molecules mimicking anticancer agents, employing DFT calculations. Model molecules will contain all the functional groups present on doxorubicin (C27H29NO11) and methotrexate (C20H22N8O5). We will thus obtain information on the mode and strength of the adsorption of the molecules on the apatite. Moreover, this model will allow us to study the effect of ion substitution, by doping of phospho-calcic apatite, on the adsorption of the molecules. This is related to the control of the kinetics of drug delivery.
In a second step, classical molecular dynamics using appropriate force fields will be used. This technique allows the treatment of much larger systems [13,14] and therefore, the study of the adsorption of anticancer molecules themselves will then be possible. The use of the data from the previous DFT calculations (geometrical parameters, charges, normal modes of vibrations) in our classical molecular dynamics model will allow us to reproduce the support/adsorbate interactions and then refine our apatite/anticancer agents' model.

(ii) Structural modification of the molecules: Quantum calculations
In the framework of the present project, state-of-the-art computational protocols will be used in terms of level of theory, mainly DFT, choosing the appropriate functionals, basis sets, environment description, and strategies to properly include weak interactions. QM calculations will be used, obviously, also to study the interactions of drugs with delivery systems, allowing us to obtain detailed information about the strength and the kind of such interactions. To this aim, specific computational tools will be adopted, such as the RDG (reduced density gradient) analysis [15,16].