Abstract:Advances in oncological research have clearly shown that beyond chemical and biological factors, mechanics and physics have a substantial role in tumor growth. The tumor mass is indeed a thermodynamical system interacting with the host tissue and its vascular and lymphatic system with which it exchanges mass and energy [1, 2]. For these reasons, over the past decade, an increasing number of mathematicians, physicists and research engineers is becoming involved in oncology research, which is nowadays, especially in the US, profoundly multidisciplinary. Within this context, mathematical modeling of tumor growth plays a key role in the development of new therapeutic strategies. A new scientific research domain is emerging: “transport oncophysics”. A mathematical model for tumor growth has been recently developed by the prospective PhD supervisor G. Sciumè et al. [3-8] (I2M-Bordeaux). In this model tumor tissue and its microenvironment are treated as a deformable porous medium permeated by three immiscible complex fluids (healthy tissue cells, cancer cells, and interstitial fluid). The proposed model is very general, and with its specificities, is currently one of the most complete tools existing in literature. Some of the constitutive equations used in this theory have been derived exploiting strong mathematical formulation analogies between the problem of tumor growth and that of flows in soils and / or geomaterials [9]. Figure 1 – Tumor growing in proximity of two healthy blood vessels http://legato-team.eu Employing recent mechanistic experimental results (e.g. [10]) and fresh data on tumor growth in confined environments (research team of P. Nassoy at Optics Institute of Aquitaine), the first objective of the PhD thesis is therefore to study and improve when necessary the consistency of adopted constitutive laws. Attention will be paid on identifying the physical and biological properties that have a major influence on tumor growth and invasion. Moreover, the PhD candidate will investigate the tradeoff between “data-driven prediction” based on machine learning algorithms, and “physics-based modelling” where tumor growth is modelled from first principles. In parallel with the validation/improvement of the cell migration model, setting up the model and associated methodologies for clinical use is the second objective of the PhD thesis. Machine learning, Bayesian inverse problems for parameter identification and model order reduction for the free boundary problem of interest will be implemented by the PhD student. This will be facilitated by the collaboration between the Legato Team led by S. Bordas (http://legato-team.eu partner team in this project) and the research team of W. Skalli and PY Rohan of the l’Institut de Biomécanique Humaine Georges Charpak (IBHGC). This work will give to the model the needed flexibility for use in personalized/precision medicine by striking the required balance between physics-based modelling and machine learning algorithms. The relevance of the developed approach will be tested on cases of tumor growth in bone tissue. Clinical and experimental data will be provided by partner university-hospital teams in Paris and Bordeaux and processed by the doctoral student jointly with a research master student at IBHGC. These data will allow characterizing material and geometry and to build the personalized FE mesh at the macroscopic level. A major issue will be to establish the correct degree of detail of the model by identifying the phenomena/factors having a predominant impact on tumor growth and those that can be disregarded. This will be done using recent work performed at the Legato Team. A multi-scale model based on a structural zoom can be also considered.

Image-based biomarkers for engineering neuroblastoma patient-specific computational models

Three-dimensional image-based mechanical modeling for predicting the response of breast cancer to neoadjuvant therapy

Personalized Radiotherapy Design for Glioblastoma: Integrating Mathematical Tumor Models, Multimodal Scans and Bayesian Inference

Image-based personalization of computational models for predicting response of high-grade glioma to chemoradiation

Towards personalized computer simulations of breast cancer treatment

Integrated Biophysical Modeling and Image Analysis: Application to Neuro-Oncology

Improving personalized tumor growth predictions using a Bayesian combination of mechanistic modeling and machine learning

Heterogeneous biomechanical/mathematical modeling of spatial prediction of glioblastoma progression using magnetic resonance imaging-based finite element method

In vivo echographic evidence of tumoral vascularization and microenvironment interactions in metastatic orthotopic human neuroblastoma xenografts

Multi-modal 3-Dimensional Visualization of Pediatric Neuroblastoma: Aiding surgical Planning Beyond Anatomical Information

Patient-specific prostate tumour growth simulation: a first step towards the digital twin

Forecasting tumor and vasculature response dynamics to radiation therapy via image based mathematical modeling

Towards patient-specific optimization of neoadjuvant treatment protocols for breast cancer based on image-guided fluid dynamics

Modelling and simulation of anisotropic growth in brain tumours through poroelasticity: A study of ventricular compression and therapeutic protocols

Comprehensive quantifications of tumour microenvironment to predict the responsiveness to immunotherapy and prognosis for paediatric neuroblastomas

Patient-specific, mechanistic models of tumor growth incorporating artificial intelligence and big data

Data Driven Tumor Growth Modelling : mechanics of multiphase systems and model reduction for personalized medicine Trade-off between machine learning and physics-based modelling in oncology

Predicting the spatio-temporal response of recurrent glioblastoma treated with rhenium-186 labelled nanoliposomes

Ensemble inversion for brain tumor growth models with mass effect

In vivo characterization of brain tumor biomechanics: magnetic resonance elastography in intracranial B16 melanoma and GL261 glioma mouse models

Abstract P1-08-31: Simbiosys tumorscope: Biophysical modeling of patient-specific response to chemotherapy