Typical experiment vs. in-cell like conditions in magnetic hyperthermia: Effects of media viscosity and agglomeration

Highlights

• Magnetic nanoparticles heating power alters when medium changes from liquid to gel.

• Inductive determination of magnetic cycles in both media shows clear differences.

• Agglomerated particles in gel present ordered structure with well defined spacing.

• SAR values for the same particles are larger in gel matrix than in liquid suspension.

• Results at different field parameters show a lower resonant frequency in the gel.

Abstract

Magnetic nanoparticles (MNPs) can be used to transform electromagnetic energy into heat in hyperthermic treatment of cancer and other thermally activated therapies. The MNPs heating efficiency depends strongly on the combination of the MNPs’ structural properties and environmental conditions. MNPs hyperthermic yield is usually studied in diluted suspensions, although, in the actual therapy, the particles end mostly aggregated and fixed into cellular structures.

In this work, the heating efficiency of low size dispersion ${\mathit{Fe}}_3{O}_4$ MNPs, defined as the Specific Absorption Rate (SAR), was studied in two conditions: liquid suspension (ferrofluid FF, typical characterization state) and gel matrix (ferrogel FG, mimicking biological application environment). The samples were characterized by TEM, ZFC-FC and SAXS. Their magnetic response to radio-frequency fields was measured by induction in order to obtain SAR values from the magnetization cycles area. 3D maps of SAR versus field amplitude and frequency were elaborated in order to compare the response of fixed and suspended MNPs. Structural characterization shows FG’s MNPs agglomerated in a crystal-like mesostructure with a well defined interparticle distance. SAR results show a clear difference of behaviour between liquid and gel matrices, with larger SAR values for the FG sample indicating a lower resonance frequency, inside the studied region, for fixed MNP. Additionally, the local maximum suggested in FG’s SAR map indicates a behaviour outside linear response regimen as expected for the applied field amplitudes.

Introduction

Magnetic nanoparticles (MNPs) are being extensively studied for their applications in biomedicine [1]. In cancer treatment, the MNPs are used as a heating agent for thermoablation and magnetic fluid hyperthermia [2]. In these therapies, the particles are introduced inside the tumor and the region is exposed to a radio-frequency electromagnetic field (RF) with frequencies around 100 kHz and amplitudes up to 15 kA/m [3]. The MNPs absorb energy from the field and release it to their surroundings as heat, producing thermal damage to the tumor [4], [5]. In order to deliver an adequate thermal dose, a key aspect for these therapies is a thorough and trustworthy knowledge of the MNPs heating efficiency. This efficiency is quantified by the Specific Absorption Rate (SAR) i.e. the amount of power the particles absorb from the field per unit mass. For a set of MNPs, the SAR value is not only determined by the particles’ properties, but also by the viscosity of the supporting medium, the interaction between particles, and the frequency f and amplitude $H_0$ of the applied field. So it is that two identical MNP assemblies supported in different media and exposed to the same RF could exhibit different SAR values. This effect has been studied by comparing the thermal dissipation for a single applied field frequency of MNPs supported in liquid with MNPs supported in hydrogel [6], glycerol [7] and gelatine [8], and for many frequencies in agar [9]. Also, it has been shown that MNPs are fixed rather strongly to the tumour tissue after injection into experimentally grown tumours in mice [8]. In all cases results indicate a noticeable diminution of the power dissipation for the fixated MNPs. This effect is generally attributed to the cancellation of Brown’s dissipation mechanism although this cancellation will provoke a SAR diminution only for frequencies larger than the resonance frequency of the sample. In this direction, a recent publication by Cabrera et al. [10] suggests that the principal effect of the internalization of MNPs by living cells is due the increase in agglomeration rather than immobilization.

The typical method for SAR determination is the calorimetric measurement of the power dissipation of MNPs in liquid suspension. This method provides a direct result from the temperature increase of the studied ferrofluid (FF) but presents several limitations for the characterization of solid and biological samples. In recent years an alternative method based on the inductive determination of the RF hysteresis loops has been developed by several research groups with very good results [7], [10], [11], [12], [13], [14].

In this work, the SAR dependencies with $H_0$ and f of magnetite MNPs ferrofluid (FF) and ferrogel (FG) are studied using RF hysteresis loop area determination by induction measurements. This method allows to perform several measurements in a short time, so it was used to construct colour maps of SAR values versus field amplitude and field frequency by sweeping through several RF generator configurations. These maps are used to compare the performance of two samples that $a\mathit{priori}$ differ only in their supporting media: the FF represents the typical and simplest media for studying MNPs, while the FG constitutes a high viscosity matrix where MNPs are fixed and usually present some degree of agglomeration. This fixed-agglomerated particle condition is similar to the final state of the MNPs in biological media after their incorporation by the cells as reported in 15].