A prediction model for magnetic particle imaging-based magnetic hyperthermia applied to a brain tumor model

Brain tumor is a global health concern at the moment. Thus far, the only treatments available are radiotherapy and chemotherapy, which have several drawbacks such as low survival rates and low treatment efficacy due to obstruction of the blood-brain barrier. Magnetic hyperthermia (MH) using magnetic nanoparticles (MNPs) is a promising non-invasive approach that has the potential for tumor treatment in deep tissues. Due to the limitations of the current drug-targeting systems, only a small proportion of the injected MNPs can be delivered to the desired area and the rest are distributed throughout the body. Thus, the application of conventional MH can lead to damage to healthy tissues.

Magnetic particle imaging (MPI)-guided treatment platform for MH is an emerging approach that can be used for spatial localization of MH to arbitrarily selected regions by using the MPI magnetic field gradient. Although the feasibility of this method has been demonstrated experimentally, a multidimensional prediction model, which is of crucial importance for treatment planning, has not yet been developed. Hence, in this study, the time dependent magnetization equation derived by Martsenyuk, Raikher, and Shliomis (which is a macroscopic equation of motion derived from the Fokker-Planck equation for particles with Brownian relaxation mechanism) and the bio-heat equations have been used to develop and investigate a three-dimensional model that predicts specific loss power (SLP), its spatio-thermal resolution (temperature distribution), and the fraction of damage in brain tumors.

Based on the simulation results, the spatio-thermal resolution in focused heating depends, in a complex manner, on several parameters ranging from MNPs properties to magnetic fields characteristics, and coils configuration. However, to achieve a high performance in focused heating, the direction and the relative amplitude of the AC magnetic heating field with respect to the magnetic field gradient are among the most important parameters that need to be optimized. The temperature distribution and fraction of the damage in a simple brain model bearing a tumor were also obtained.

The complexity in the relationship between the MNPs properties and fields parameter imposes a trade-off between the heating efficiency of MNPs and the accuracy (resolution) of the focused heating. Therefore, the system configuration and field parameters should be chosen carefully for each specific treatment scenario. In future, the results of the model are expected to lead to the development of an MPI-guided MH treatment platform for brain tumor therapy. However, for more accurate quantitative results in such a platform, a magnetization dynamics model that takes into account coupled Néel-Brownian relaxation mechanism in the MNPs should be developed.