Magnetic hyperthermia has a significant potential to be a new breakthrough for cancer treatment. The simple concept of nanoparticle-induced heating by the application of an alternating magnetic field has attracted much attention, as it allows the local heating of cancer cells, which are considered more susceptible to hyperthermia than healthy cells, while avoiding the side effects of traditional hyperthermia. Despite the potential of this therapeutic approach, the idea that local heating effects due to the application of alternating magnetic fields on magnetic nanoparticle-loaded cancer cells can be used as a treatment is controversial. Several studies indicate that the heating capacity of magnetic nanoparticles is largely reduced in the cellular environment due to increased viscosity, aggregation and dipolar interactions. However, an increasing number of studies, both in vitro and in vivo, show evidence of successful magnetic hyperthermia treatment on several different types of cancer cells. This apparent contradiction might be due to the use of different experimental conditions. Here we analyze the effects of several parameters on the cytotoxic efficiency of magnetic nanoparticles as heat inductors under an alternating magnetic field. Our results indicate that cell-nanoparticle interaction reduces the cytotoxic effects of magnetic hyperthermia, independently of nanoparticle coating and core size, the cell line used and the subcellular localization of nanoparticles. However, there seems to occur a synergistic effect between the application of an external source of heat and the presence of magnetic nanoparticles, leading to higher toxicities than those induced by heat alone, or the accumulation of nanoparticles within cells.

The aggregation processes of magnetic nanoparticles in biosystems are analysed by comparing the magnetic properties of three systems with different spatial distributions of the nanoparticles. The first one is iron oxide nanoparticles (NPs) of 14 nm synthesized by coprecipitation with two coatings, (3-aminopropyl)trimethoxysilane (APS) and dimercaptosuccinic acid (DMSA). The second one is liposomes with encapsulated nanoparticles, which have different configurations depending on the NP coating (NPs attached to the liposome surface or encapsulated in its aqueous volume). The last system consists of two cell lines (Pan02 and Jurkat) incubated with the NPs. Dynamic magnetic behaviour (AC) was analysed in liquid samples, maintaining their colloidal properties, while quasi-static (DC) magnetic measurements were performed on lyophilised samples. AC measurements provide a direct method for determining the effect of the environment on the magnetization relaxation of nanoparticles. Thus, the imaginary (χ'') component shifts to lower frequencies as the aggregation state increases from free nanoparticles to those attached or embedded into liposomes in cell culture media and more pronounced when internalized by the cells. DC magnetization curves show no degradation of the NPs after interaction with biosystems in the analysed timescale. However, the blocking temperature is shifted to higher temperatures for the nanoparticles in contact with the cells, regardless of the location, the incubation time, the cell line and the nanoparticle coating, supporting AC susceptibility data. These results indicate that the simple fact of being in contact with the cells makes the nanoparticles aggregate in a non-controlled way, which is not the same kind of aggregation caused by the contact with the cell medium nor inside liposomes.