Towards a Digital Twin of Magnetic Nanoparticle-Mediated Cancer Therapy
Please login to view abstract download link
To overcome the limitations of conventional cancer therapies, magnetic nanoparticle-based strategies have emerged as a promising avenue for specific tumor targeting. A particularly innovative approach involves loading cells (e.g., T cells for immunotherapy) with magnetic nanoparticles and guiding them to the tumor site via an external magnetic field. While such magnetically enhanced cellular therapies have shown impressive clinical responses, the critical parameters governing the successful accumulation of nanoparticles or magnetized cells in tumors remain poorly understood. We, therefore, present the foundational steps and key challenges in developing a digital twin for magnetic nanoparticle-mediated cancer therapy. This includes a computational framework that integrates a physics-based model of magnetic nanoparticle transport under an external field with a multiphysics tumor growth model based on porous media theory. Furthermore, we demonstrate a rigorous methodology to 1) assess model sensitivity to input parameters via global sensitivity analysis, and 2) calibrate the model consistently against experimental data using Bayesian inference. We argue that the iterative integration of such computational models with in vitro and in vivo experiments in a continuous feedback loop is essential. This synergistic approach is the key to realizing a predictive digital twin, which will ultimately enable the design of optimized, personalized treatment strategies.
