Mathematical modelling and optimisation of organ-on-a-chip in vitro systems

在临床前药物开发中，充分预测药物毒性的一个基本特征是使用细胞体外模型，尽可能地概括人体组织的生理学。包括生理上真实的流体流动的三维细胞系统对于提供机械生物学反应和细胞正确功能所需的适当剪切应力是重要的。它们还可以提供药物、营养物质和废物化合物的运输和再循环，以及细胞因子和趋化因子等信号分子，这些信号分子驱动细胞间的通信，这对体外细胞的生理功能至关重要。这些溶质的分布与理解细胞功能有关，可用于增强毒理学研究中的药代动力学和细胞-细胞相互作用以及信号模型。一个特别的焦点是肝细胞的器官芯片模型，因为肝毒性是药物开发临床失败的主要原因。这些可以扩展到包括多种细胞类型或多器官系统（例如免疫细胞，肠道细胞），以纳入与药物作用机制相关的进一步相互作用。这些系统的数学建模不仅有助于理解和改善体外模型的生理相关性，而且还将使相关实验设置优化，最重要的是，使药物开发过程中毒性的定量预测成为可能。目的和目的构建一系列器官芯片系统的流体流动和溶质运输的机械数学模型，并结合细胞功能的相关模型（如代谢、免疫介导的效应靶毒性、细胞因子释放-细胞间通讯）。获得流体流动、剪切应力和浓度分布的定量预测，并与实验获得的结果进行比较。确定体外系统的最佳设计和操作条件（流速，支架特性，系统几何形状），以便尽可能地匹配细胞所经历的体内环境和/或优化系统的性能，作为毒性评估的工具。通过对流体流动的理解，告知并优化实验设计和药代动力学模型。了解流体动力负荷对细胞功能的影响。获得可以应用和适应各种微流体系统的一般数学框架，其中流体力学的先进理解可以通过复杂流体流动和细胞功能之间的相互作用提供基本见解，为药物发现过程提供信息。研究方法的新颖性研究方法将包括机械数学建模、分析和计算机计算，并结合罗氏进行的实验研究。数学模型将结合流体动力学建模的思想（Navier-Stokes， Darcy/Brinkman方程，多相流，反应-平流-扩散方程），适用于系统的不同组成部分。该模型将通过数值技术（如有限差分，有限元和频谱方法，使用商业软件）和利用不同时间/长度尺度（如线性和非线性稳定性理论，正则和奇异摄动理论，多尺度分析，参数空间中的极限情况）获得的简化模型的分析方法的组合进行研究。数据将从罗氏进行的实验研究和成像中获得。

An essential feature of adequate prediction of drug toxicity in preclinical pharmaceutical development is the use of cellular in vitro models that recapitulate the physiology of human tissues as closely as possible. 3D cellular systems that include physiologically realistic fluid flow are important in providing appropriate shear stresses required for mechanobiological responses and correct function of cells. They can also provide the transport and recirculation of drugs, nutrients and waste compounds, as well as signalling molecules such as cytokines and chemokines, which drive cell-cell communication that is critical for physiological functioning of cells in-vitro. The distribution of such solutes is relevant to understanding cellular function and can be used to enhance pharmacokinetic and cell-cell interaction and signalling models in toxicology studies. A specific focus is organ-on-a-chip models of liver cells, since hepatotoxicity is a major cause of clinical failure in drug development. These can be extended to include multiple cell types or multi organ systems (e.g. immune cells, gut cells) to incorporate further interactions relevant to the drug's mechanism of action. Mathematical modelling of such systems not only helps to understand and improve the physiological relevance of in vitro models, but will also enable the optimisation of relevant experimental settings, and most importantly, enable quantitative predictions regarding toxicity in the drug development process.Aims and ObjectivesConstruct mechanistic mathematical models for fluid flow and solute transport for a range of organ-on-a-chip systems, coupled to relevant models of cellular function (e.g. metabolism, immune mediated effector-target toxicity, cytokine release cell-cell communication).Obtain quantitative predictions of fluid flow, shear stresses, and concentration distributions, and compare with results obtained experimentally.Determine optimal design and operating conditions of the in vitro system (flow rates, scaffold properties, system geometry) in order to match the in vivo environment experienced by cells as closely as possible and/or optimise the performance of the systems as a tool for toxicity assessments. Inform and optimise experimental design and pharmacokinetic modelling through understanding of fluid flow.Understand impact of fluid dynamical load on cellular function.Obtain a general mathematical framework that can be applied and adapted to a variety of microfluidic systems, where advanced understanding of fluid mechanics can provide fundamental insights through the interaction between complex fluid flows and cell function, informing the drug discovery process.Novelty of Research MethodologyResearch methodology will include mechanistic mathematical modelling, analysis and in silico computation, in combination with experimental studies performed at Roche. The mathematical model will incorporate a combination of ideas from fluid dynamic modelling (Navier-Stokes, Darcy/Brinkman equations, multiphase flows, reaction-advection-diffusion equations) which apply to different components of the system.The model will be investigated through a combination of numerical techniques (e.g. finite-difference, finite element and spectral methods, use of commercial software) and analytical approaches on reduced models obtained by exploiting different time/length scales (e.g. linear and nonlinear stability theory, regular and singular perturbation theory, multiple-scales analysis, limiting cases in parameter space).Data will be obtained from experimental studies and imaging performed at Roche.