Computational modeling of multiphase transport in physiologically realistic solid tumor vasculature and intratumoral domains

Clinical diagnosis and targeted drug delivery assessments for dense tumors (e.g., in pancreatic cancer) can benefit from a quantitative framework that can project intratumoral plasma uptake from simple inputs, e.g., medical scans of the tumor. To that end, we are building a first-principles mechanics-based model that incorporates features such as the extracellular packing fraction in the tumor and the outer vasculature shapes. We have numerically modeled Eulerian multiphase blood transport in vessels reconstructed from high-resolution scans of human pancreatic tumors implanted in mice. Both red and white blood cells, along with plasma, are the different phases considered in the viscous-laminar transient simulations, which also incorporate electrohydrodynamic and pulsatile effects to capture the essential biological realism. Subsequently, we have used the plasma flow parameters at the fenestra openings to the tumor, to numerically model plasma perfusion through 2D idealized intratumoral domains that bear packing fractions similar to real tumors. Our current findings show inverse correlation between plasma percolation rates and the intratumoral diffusion distances.