Simulation Based Strategies for Clinical Translation of Magnetic Nanoparticle Hyperthermia
Johns Hopkins University
Magnetic nanoparticles have gained significant importance in the recent past for their use in biomedical applications such as drug delivery, imaging, diagnosis, and therapy. Magnetic nanoparticle hyperthermia is the selective heating of tumor tissue using magnetic nanoparticles which generate heat when exposed to an alternating magnetic field. It is a minimally invasive method which can cause effective and localized tumor thermal damage. The challenge to achieve consistent heating with this modality is the variable distribution upon delivery, which results in variable heat distribution in the tumor and surrounding normal tissue. In this thesis, using computational methods we explore optimization strategies to modulate magnetic field amplitude using limited temperature feedback to achieve clinically effective thermal dose in tumor and minimize healthy tissue damage. The magnetic field amplitude is modulated by using a Proportional-Integral-Derivative (PID) controller based on temperature feedback from tumor-healthy tissue boundary. We consider nanoparticle distributions obtained from animal studies and idealized mathematical constructs. Two and three dimensional (2D & 3D) models of tumor and healthy tissue were considered. Temperature effects on perfusion were considered. Results of thermal damage, temperature distributions and thermal dose obtained from modulated power heating were then compared to constant power heating. It is shown that controlling the tumor-healthy tissue boundary temperature by modulating the heating power of the nanoparticles can compensate for variable nanoparticle distributions to deliver effective treatment. The strategy was then implemented in mouse models of liver cancer. Two nanoparticle distributions were generated by using two injection methods. It was shown that the temperature at the tumor-healthy tissue boundary can be consistently controlled for the two nanoparticle distributions. The challenges associated with implementation of our proposed strategy have been identified and future steps for further accurate testing have been presented. Another challenge for magnetic nanoparticle hyperthermia is the onset of eddy current heating when the treatment modality is applied to tumors in large organs. Monitoring of eddy current heating in in vivo studies is challenging. Hence, we developed a computational tool which couples thermal and electromagnetic modeling to predict the temperatures achieved due to eddy current heating. The model was verified with the analytical solution and validated with gel phantom experiments. We then implemented it to generate 3D liver model from computed tomography (CT) images of rabbit liver. The temperatures attained due to eddy current heating from exposure to alternating magnetic fields were calculated to demonstrate the utility of the model in estimating temperature during magnetic nanoparticle hyperthermia of large organs. In the last chapter, we characterized the thermal and magnetic properties of dual contrast nanoparticle formulations used in image guided thermal therapy of liver cancer. Dual contrast nanoparticle formulations are magnetic iron oxide nanoparticles combined with lipiodol. The heating potential of these lipiodol nanoparticle formulations was extensively characterized by measuring their thermal properties at fixed frequency with different magnetic field amplitudes. These were then compared to original aqueous formulations for assessing the differences between both the formulations. Bulk magnetic properties of both the formulations was measured and compared. It is observed that when nanoparticles are mixed with lipiodol, the specific loss power of these particles is reduced. These results highlight the importance of evaluating the heating performance of new nanoparticle formulations.
heat transfer, bioheat transfer, cancer, hyperthermia, nanoparticles, PID control, lipiodol, magnetic nanoparticles, specific power loss, thermal dose, perfusion modeling, eddy current heating, finite elements, simulation, thermal damage, thermal medicine