(F-215) Development of in Silico Model of Pancreatic Tissue

Introduction::

Mathematical and computational modelling of cancerous tissue represents a means by which mechanical properties of tissues, efficacy of certain drugs, as well as immune response to cancer can be simulated without the use of animal testing to better inform in vivo preclinical trials, as well as to gain viscoelastic properties of cancerous tissues without needed a biospecimen of a specific patient.

Materials and Methods:: Tumor tissue consists of stroma and extracellular matrix (ECM), and cancer cells. Here, we take the first step to model tumor issue where the components are distinct with their mechanical properties. Figure 1 shows our model of that tumor which has been generated in LAMMPS, a classical molecular dynamics open-source software. LAMMPS is being utilized in this study to conduct a coarse-grained modelling of pancreatic tissue microenvironment. The assumption of a coarse-grained modelling system is that objects such as proteins, or larger objects such as drug particles or even cells can be represented as a discrete set of points which interact in pairwise interactions to approximate the mechanical properties of the object in the mesoscopic scale. The mesoscopic scale is typically defined to be the range between 10 and 1000 nm where the acuity of the simulation is greater than that of continuous scale, but rougher than that of a simulation concerned with the level of detail to the chemical bonding level.

Results, Conclusions, and Discussions:: Initial results show that it is possible to replicate the mechanical properties of cells from existing literature. It is then possible to tune the parameters defined in the software to distinguish different cells or drug particles not only from defined geometries, but also in terms of stiffness, resistance to volume change, etc. We have determined the relation between stiffness of cell types e.g., red blood cells and potential functions defined to model the intracomponent interactions between the points forming each component. This relation has been validated for red blood cells against a recent computational study.

This research project is under current development with progress being made in accurately modelling individual objects, such as cells or drug particles. Future plans are to run simulations with layers of particles, representing tissue layers of a tumor microenvironment, while simulating diffusion of drug particles, using random placement of drug particles to study the penetration depth and adherence of the drugs, given their mechanical properties. Other areas of future study are to research the effect that cells of different mechanical properties (cancer cells) affect the overall tissue properties.

Acknowledgements (Optional): :

References (Optional): : Scott Diamond, et al. A parallel fluid-solid coupling model using LAMMPS and
Palabos based on the immersed boundary method. March 2018