External Basic Hyperthermia Devices for Preclinical Studies in Small Animals

Simple Summary The application of mild hyperthermia can be beneficial for solid tumor treatment by induction of sublethal effects on a tissue- and cellular level. When designing a hyperthermia experiment, several factors should be taken into consideration. In this review, multiple elementary hyperthermia devices are described in detail to aid standardization of treatment design. Abstract Preclinical studies have shown that application of mild hyperthermia (40–43 °C) is a promising adjuvant to solid tumor treatment. To improve preclinical testing, enhance reproducibility, and allow comparison of the obtained results, it is crucial to have standardization of the available methods. Reproducibility of methods in and between research groups on the same techniques is crucial to have a better prediction of the clinical outcome and to improve new treatment strategies (for instance with heat-sensitive nanoparticles). Here we provide a preclinically oriented review on the use and applicability of basic hyperthermia systems available for solid tumor thermal treatment in small animals. The complexity of these techniques ranges from a simple, low-cost water bath approach, irradiation with light or lasers, to advanced ultrasound and capacitive heating devices.


Introduction
Hyperthermia (HT) is a therapeutic modality in which tissue temperature is elevated above physiological temperature for a predefined period of time. Application of mild hyperthermia (40-43 • C) is able to induce sublethal effects to the target region, which can induce beneficial mechanisms on both a tissue-and cellular level [1,2]. Clinical hyperthermia can be divided into three categories: local-, partial-, or whole body hyperthermia [3]. To study the application, effects in a treatment setting, and mechanisms involved in HT, different methods have been developed. The selection of the most suitable heating method depends on the research question, (pre)clinical setting, and most importantly on the complexity of the heating method needed.
Hyperthermia is usually administered as an adjuvant therapy to improve the therapeutic efficacy of existing radio-and/or chemotherapies. The main focus of this review is devices inducing mild, local temperature elevation in the treatment of small animals bearing solid tumors. Various methods have been developed, varying in their way of energy transfer, heating volume, site of application, complexity, and costs [4]. In the following sections, the general considerations for designing a preclinical heating study are outlined. Subsequently, the available hyperthermia techniques are described in detail. These range from simple setups such as water baths, cold light sources, and near-infrared lasers, to advanced focused ultrasound and capacitive hyperthermia devices. This review provides an outline on devices and adaptations thereof, which can be used for application of mild HT in small animals, and helps to identify the optimal setup depending on the research question.

Temperature Monitoring
The normal rectal temperature in small animals varies slightly for mice (36.5-38.0 • C), rats (37.5-38.5 • C), and hamsters (37.0-38.0 • C) [23]. The rectal temperature should not rise above 39 • C, nor manifest a maximum variation of 0.5 • C during the heating studies [24][25][26]. Thermoregulation can be maintained during the experiment with the use of heater air, -plates, -pads, or reflective foils [27,28]. Some studies have simultaneously applied cooling of the top surface by gently blowing room temperature air over the animal. For instance, studies applying water bath hyperthermia have incorporated air-cooling to avoid increase of physiological core temperature due to thermal conduction [22,29,30].
Preliminary heating experiments should be performed to determine heat distribution for each system at various settings. In order to obtain the general heating pattern, the temperature should be monitored at multiple locations in the tumor center and -periphery [29,31,32]. In the past, most of the preclinical HT studies measured the intratumoral temperature with the use of needle-type thermocouples. However, the invasiveness of this method should be taken into account, as probe insertion leads to parenchymal disruption, and may create artefacts in tumor circulation [33]. Another factor to take into consideration for temperature monitoring with thermocouples is the potential radiofrequency interaction. Application of capacitive hyperthermia may cause interaction between the conductive metallic components of the thermocouple and the electromagnetic field, resulting in measurement artefacts [34,35]. Additionally, self-heating of the metallic-based thermocouples could eventually cause overestimation of the tissue temperature [36].
In a more advanced setup, hyperthermia profiles could be determined using noninvasive real-time imaging techniques in the form of infrared (IR)-or magnetic resonance (MR) thermometry [37,38]. IR thermometry is able to provide 2D temperature maps, as only superficially emitted infrared energy is converted into temperature profiles [39,40]. Noninvasive 3D thermal monitoring can be obtained using proton resonance frequency shift (PRFS) methods. PRFS is based on the acquisition of phase distribution images through gradient echo (GRE) sequences. Reference-based temperature maps can be generated in real time by subtracting the background image acquired before hyperthermia application from the images obtained during the experiment. However, it is important to consider that this technique only monitors the change in temperature and does not provide absolute temperatures [41]. In addition, both non-invasive imaging techniques may not provide the same accuracy as the thermometry probes due to tissue dependence and susceptibility to motion artifacts [40,42].

Water Bath
Application of hyperthermia by immersion of the target area into a temperaturecontrolled water bath is a low-complexity technique (Figure 2A). In the past decades, various groups have studied the effect of this approach on mice, rats, and hamsters (Table 1). In order to minimize harmful effects to non-targeted tissue regions, there are factors to consider for standardization in future experiments.
In animal experiments involving the water bath, the animal receives a subcutaneous tumor suspension or fragment most frequently on one or both hind limbs, but water bath heating on the flank or mammary fat pad is also an option (Table 1). Both minor and major precautions aid the protection of healthy tissue surrounding the tumor during heat application. For instance, application of Vaseline cream to the tumor border forms a protective barrier, which protects non-cancerous tissues from heating damage and, therefore, prevents possible skin burns [43][44][45]. Others have covered the to-be immersed leg with a thin plastic bag in order to prevent excessive water absorption. On the long term, this will prevent formation of limb edema [32,46,47]. Another possibility is the use of Tesa™ tape to fix the animal in the correct position. The tail, leg, or body can be loosely taped to a jig or device in order to ensure a similar maximum depth of immersion for every experiment [24,32,33,48]. The aligned temperature probes were placed in the tumor center at different depths in relation to the water surface. Image B is adapted from Nishimura et al. [29].
In animal experiments involving the water bath, the animal receives a subcutaneous tumor suspension or fragment most frequently on one or both hind limbs, but water bath heating on the flank or mammary fat pad is also an option (Table 1). Both minor and major precautions aid the protection of healthy tissue surrounding the tumor during heat application. For instance, application of Vaseline cream to the tumor border forms a protective barrier, which protects non-cancerous tissues from heating damage and, therefore, prevents possible skin burns [43][44][45]. Others have covered the to-be immersed leg with a thin plastic bag in order to prevent excessive water absorption. On the long term, this will prevent formation of limb edema [32,46,47]. Another possibility is the use of Tesa™ tape to fix the animal in the correct position. The tail, leg, or body can be loosely taped to a jig or device in order to ensure a similar maximum depth of immersion for every experiment [24,32,33,48]. Some studies have ensured full extension of the immersed leg via attachment of a sinker (approximately 4 g) to the plantar surface of the foot [26,29,33,49].
The animal is subsequently placed on a platform, which should be adapted to reduce thermal conduction from the water bath and simultaneously ensure complete immersion of the tumor-bearing leg. The stage should be covered with an insulating material, such as polystyrene or plastic, in order to minimize an increase in body temperature due to heat radiation [15,26,50,51]. This layer should be thin enough to establish complete immersion of the malignant tissue, as the creation of an air-tissue interface will lead to convective heat loss [22]. Combining these modifications reduces the development of hyperthermia related side effects, such as edema or hemorrhagic necrosis, and allows for complete recovery within 7 days post-treatment [32].
The usefulness of this method lies in its user-friendliness and the fact that any water bath can easily be converted to serve as a hyperthermia device. However, the main drawback of this technique is its range of application, as it is only suitable for models studying superficial (subcutaneous-and intramuscular) tumors. Therefore, the system only mimics the clinical setting for certain cancers, such as melanoma.
An aspect to consider when applying water bath hyperthermia is the non-specific temperature elevation. While other HT devices described in this paper focus on targeted treatment, water bath heating is rather non-specific, affecting an area significantly larger than only the tumor. This creates a different environment, accommodating more than just a solid tumor and the aberrant vasculature. In non-cancerous tissue, the induction of thermal stress results in a thermoregulatory response, which may subsequently result in an increase in local perfusion [52]. Previously, Tungjitkusolmun et al. and Ware et al. observed that dissipation of thermal energy occurs depending on the distance between blood vessels and the tumor [53,54]. As blood flow in larger vessels can act as a heat sink, The animal is subsequently placed on a platform, which should be adapted to reduce thermal conduction from the water bath and simultaneously ensure complete immersion of the tumor-bearing leg. The stage should be covered with an insulating material, such as polystyrene or plastic, in order to minimize an increase in body temperature due to heat radiation [15,26,50,51]. This layer should be thin enough to establish complete immersion of the malignant tissue, as the creation of an air-tissue interface will lead to convective heat loss [22]. Combining these modifications reduces the development of hyperthermia related side effects, such as edema or hemorrhagic necrosis, and allows for complete recovery within 7 days post-treatment [32].
The usefulness of this method lies in its user-friendliness and the fact that any water bath can easily be converted to serve as a hyperthermia device. However, the main drawback of this technique is its range of application, as it is only suitable for models studying superficial (subcutaneous-and intramuscular) tumors. Therefore, the system only mimics the clinical setting for certain cancers, such as melanoma.
An aspect to consider when applying water bath hyperthermia is the non-specific temperature elevation. While other HT devices described in this paper focus on targeted treatment, water bath heating is rather non-specific, affecting an area significantly larger than only the tumor. This creates a different environment, accommodating more than just a solid tumor and the aberrant vasculature. In non-cancerous tissue, the induction of thermal stress results in a thermoregulatory response, which may subsequently result in an increase in local perfusion [52]. Previously, Tungjitkusolmun et al. and Ware et al. observed that dissipation of thermal energy occurs depending on the distance between blood vessels and the tumor [53,54]. As blood flow in larger vessels can act as a heat sink, the femoral vasculature in the water bath model may contribute to thermal efflux [52]. Therefore, temperatures in healthy tissue of the hind limb may rise above physiological temperature during treatment, but as long as mild hyperthermic temperatures are applied no permanent damages are expected to occur. This is in contrast to tumor tissue, as its poor circulation has created a hostile micro-environment due to both structural and functional abnormalities of the vasculature [55]. Therefore, the tumor is unable to respond to thermal stress in a similar manner as healthy tissue. As the heat may not be able to diffuse in order to restore homeostasis, the TME is more vulnerable to heat damage [56].
The application of water bath hyperthermia may also result in non-homogeneous heat distribution. As mentioned in Section 2.3, assessment of tumor temperature gradient can be obtained by placement of invasive tissue probes at various depths relative to the tumor surface. This can be either central (±9 mm), lateral (±6 mm), or peripheral (±3 mm) [41].
As the heat source is located externally, it is plausible that the highest temperature is always measured in the periphery of the tumor and that temperature decreases with increasing tissue depth [41]. As shown by Nishimura et al. and Masunaga et al., temperatures at the tumor center equilibrated within 3-4 min after complete immersion, and remained at 0.2-0.3 • C below the water bath temperature throughout the study [29,30]. The study by Nishimura et al. demonstrated that the depth of immersion is also of importance, as stabilized tumor center temperatures are shown to be 0.1 • C higher at the region most distant from the water surface [29] ( Figure 2B). The temperature distribution is dependent on the diameter and depth of the subcutaneous tumor. A study by Willerding et al. showed an increased variation among temperatures at the different probe locations in large tumors exposed to water bath heating in comparison to small tumors [41]. However, the accuracy and precision of probe placement may be low, and temperature values divergent, due to tissue disruption [51]. Therefore, homogeneous heat distribution is thought to occur most optimally in small, shallow tumors located on the distal lower limb or protruding tissue.
Lastly, the integration of either intravital imaging or magnetic resonance imaging (MRI) thermometry could be hampered due to the presence of a large water bolus. Image acquisition is rather impractical as the working distance of the water-dipping lens should attain to the complete immersion depth of the tumor tissue ( Figure 2B). MR imaging and -thermometry on the other hand could benefit from presence of the water bolus by homogenizing the magnetic field, called "shimming". Sumser et al. observed that for hyperthermia purposes, the water bolus fluid should be spiked with Fe 3 O 4 nanoparticles in order to improve the signal-to-noise ratio [57]. This is in contrast to standard water baths, which frequently use demineralized water.  3 if only measurements were provided, using the following formula: V = 0.5 (A·B 2 ). Estimates are indicated by "~".

Cold Light Source (CLS)
Cold light sources are devices that are used to heat tissues, but cause limited superficial heat deposition due to utilization of filtered light, because of which the designation "cold" is used. CLS devices are based on incandescent light bulbs, indicating that the emission of light is caused by heating of the internal filament [73]. A distinction can be made between the use of a simple halogen lamp, or a more advanced water-filtered infrared-A (wIRA) radiation system (Table 2). Standard halogen lamps emit infrared (IR) radiation, which generates heat via a combination of infrared-A (IR-A, λ = 760-1400 nm), infrared-B (IR-B, λ = 1400-3000 nm), and infrared-C (IR-C, λ = 3000 nm-1 mm) [74].
Emission throughout the entire infrared spectrum leads to heat absorption mostly in the superficial skin layers, as this region has the highest water content [75]. Subsequently this may lead to undesired side effects varying from painful sensations to irreversible tissue damage, depending on factors such as treatment duration and intensity [76]. Therapeutically relevant heating can be obtained by the use of a series of band-pass filters in order to emit light only of the desired wavelength [77]. A more sophisticated solution is the addition of a water filter to the setup. This eliminates the presence of both IR-B and IR-C, leading to a higher transmission into the skin and therefore an increase of the maximum tolerable radiation power [74,75,78,79]. The application of unfiltered heat radiation with the use of a standard halogen lamp system (EFR type) is still possible, with the limitation of the maximum intratumoral temperature being set to 41 • C [80]. Exceeding this temperature leads to the aforementioned skin damage [80]. A cold light source, for instance, the KL series from SCHOTT, Mainz, Germany, can easily be adapted for hyperthermic treatment by the removal of the glass filter.
To establish the in vivo tumor model, the animal receives a subcutaneous tumor suspension or fragment on a strategic location on the body ( Table 2). The location of the superficial tumor is adaptable and dependent on the study aims, as the motility of the fiber-optic light guide(s) allow(s) for freedom of positioning of both animal and tumor ( Figure 3A). In general, the most preferable tumor site is either back, hind leg, or one of the hind feet. Prior to start of the treatment, the animal is anesthetized and subsequently placed into the correct position and immobilized using elastic adhesive bandage (Vet Wrap ® ) [28]. In order to minimize heat application to the surrounding tissues, tumor boundaries are masked using white cotton wool, surgical swabs, or polystyrene [41,80,81]. However, it should be noted that gauze-style dressings do not completely avoid IR radiation as diffuse transmission still occurs through the weave pattern [82]. Tumor tissue is heated using a fiberoptic light source. The design of the applicator consists of a flexible light guide, which facilitates the exact placement of the light beam on the tumor area ( Figure 3B). The use of multiple guides ensures a uniform heating pattern in larger tumors [41,80]. The intensity of the lamp can be adjusted to the desired input using an adjustable power source (variac) [81,83] ( Figure 3C).
The use of CLS for the application of local hyperthermia shows promise due to the ease of assembly, steering, and alteration of intensity. However, only a limited number of groups have published this technique in the past decades (Table 2), including experimental tumor treatment in larger animals such as piglets and equines (not included in this overview) [84,85]. Hyperthermia using incandescent light can be applied to a more defined target region in comparison with full immersion in a water bath, but at the same time the application of this technique is still limited to superficial skin malignancies. Upon penetration into the skin, wave attenuation occurs due to absorption and scattering. In the case of unfiltered heat radiation, the penetration depth does not exceed several millimeters. This results in the formation of a heat gradient, with high absorption in superficial layers and deposition of sub-cytotoxic heat levels at deeper tissue regions [86]. However, it is important to realize that the cold light source profile follows black-body radiation. This means that temperature is inversely correlated with the wavelength of emitted light, for example, emitted light changes from a faint red glow to white, and, finally, into a blue color with increasing temperature [87,88].
In order to generate homogeneous heating in large tumors, application of mild hyperthermia using wIRA was suggested [89]. Previous studies have observed that a change in wavelength, from 500 to 700 nm, leads to a significant reduction in optical absorbance. Therefore, penetration depth will be enhanced, resulting in therapeutically relevant tissue heating up to a depth of 5 mm [76]. Radiative penetration reaches depths up to 10 mm for both halogen and wIRA systems; however, these may be subtherapeutic heating levels.
time the application of this technique is still limited to superficial skin malignancies. Upon penetration into the skin, wave attenuation occurs due to absorption and scattering. In the case of unfiltered heat radiation, the penetration depth does not exceed several millimeters. This results in the formation of a heat gradient, with high absorption in superficial layers and deposition of sub-cytotoxic heat levels at deeper tissue regions [86]. However, it is important to realize that the cold light source profile follows black-body radiation. This means that temperature is inversely correlated with the wavelength of emitted light, for example, emitted light changes from a faint red glow to white, and, finally, into a blue color with increasing temperature [87,88].
In order to generate homogeneous heating in large tumors, application of mild hyperthermia using wIRA was suggested [89]. Previous studies have observed that a change in wavelength, from 500 to 700 nm, leads to a significant reduction in optical absorbance. Therefore, penetration depth will be enhanced, resulting in therapeutically relevant tissue heating up to a depth of 5 mm [76]. Radiative penetration reaches depths up to 10 mm for both halogen and wIRA systems; however, these may be subtherapeutic heating levels.  The heat application through one or multiple flexible fiber optic light guides allows for relatively free positioning of the subcutaneous tumor location. The healthy tissue can be shielded from illumination by placement of, for instance, cotton wool; (B) the light source is placed above the target area without interference of any medium. The geometry of heat application is dependent on the width of the light beam and the distance of the light source in relation to the tumor; (C) the water-filtered infrared-A (wIRA) radiation system eliminates the presence of both IR-B and IR-C due to the water filter. This provides a light beam with a high penetration depth. Image C is adapted from Kelleher et al. and Vaupel et al. [76,90]. Table 2. Overview of the preclinical studies on application of local hyperthermia using a cold light source sorted by year of publication and animal model. A distinction is made between the halogen-and wIRA-based CLS systems.

Near-Infrared (NIR) Laser Light
Laser hyperthermia is based on the conversion of electrical energy into light energy, which is subsequently able to interact with tissues to produce heat [92]. The light emitted by lasers is monochromatic, directional, and coherent. It can be produced in specific wavelengths, varying from visible-to infrared light (λ = 400-3000 nm) [93]. These properties define the extent of tissue penetration and result in predictable temperature profiles [28]. Near-infrared (NIR) frequencies (λ = 750-3000 nm) are able to deliver laser light with high spatial precision over long distances without significant loss of energy [92,94]. As with the cold-light source, there is a window for optimal tissue penetration, which is based on the effective attenuation length. While light absorption depends on various cell-and tissue characteristics, wavelengths of approximately 800 nm are most frequently applied in preclinical studies (Table 3). Showing a penetration depth of up to 10 mm, with minimized excitation of biomolecules and water heating [95][96][97]. However, it should be taken into account that hemoglobin is a dominant background absorber within the 700-900 nm range [98]. This is especially relevant for wellvascularized tumors or malignancies located near major blood vessels, as treatment planning may require parameter alteration to account for heat dissipation. Wavelengths below 700 nm show diminished penetration, while water absorption is more pronounced at wavelengths exceeding 1000 nm [95,96]. The wavelengths selected for preclinical laser-based heating fall between 763 and 1064 nm (Table 3).
Another factor to take into consideration is the Gaussian intensity distribution of the laser beam [99]. When applying laser light using a bare-tipped optical fiber, the directional beam may create a hot spot in the center of the target area. Therefore, there may be non-homogeneous heating, as the tissue located at the boundaries of the beam receives a lower light dose in comparison with tissue in the beam center. Manipulation of the laser beam to establish focalized treatment occurs either through the use of multiple lenses and prisms or with the use of a custom-built illuminator. The diameter of the laser beam can be adjusted to the tumor size by varying the distance between the lenses and the prism [41,100] ( Figure 4C). The use of an illuminator provides homogeneous light distribution by passing light through three aligned chambers lined with reflective material ( Figure 4B). The size of the beam hitting the tissue surface depends on the size of the exit port in the middle chamber [101][102][103][104]. Upon reaching the tumor tissue, continuous heating occurs as the laser beam will initially heat the upper layer, which gradually spreads to the underlying tissue layers 2-10 mm deep [95][96][97]. In order to establish mild hyperthermia, the output energy of the laser can be adapted by either changing the beam size (mm) or the laser intensity (W/cm 2 ).
In animal experiments involving the NIR laser light, the animal receives a subcutaneous tumor suspension or fragment, commonly in the flank, mammary fat pad, or hind limb (Table 3). After application of anesthesia, the treatment site is prepared by shaving and subsequent cleaning with antiseptics. It is possible to opt for either systemic or topical analgesia to alleviate any pain response due to the stimulus [105]. The animal is placed on the stage in an optimal position and immobilized ( Figure 4A). During laser-light treatment, both power (W/cm 2 ) and pulse interval (s) were manually adjusted in order to ensure target temperature with minimal fluctuations. As mentioned in the general considerations, a non-invasive method to determine the real-time hyperthermia profiles is MRT. After reaching steady state temperatures, the average tumor surface temperature can be monitored during laser light irradiation. In order to ensure MR compatibility, studies have designed holders, mounts and other features based on non-magnetic materials such as polyethylene and glass [41,100,106]. middle chamber [101][102][103][104]. Upon reaching the tumor tissue, continuous heating occurs as the laser beam will initially heat the upper layer, which gradually spreads to the underlying tissue layers 2-10 mm deep [95][96][97]. In order to establish mild hyperthermia, the output energy of the laser can be adapted by either changing the beam size (mm) or the laser intensity (W/cm 2 ). In animal experiments involving the NIR laser light, the animal receives a subcutaneous tumor suspension or fragment, commonly in the flank, mammary fat pad, or hind limb (Table 3). After application of anesthesia, the treatment site is prepared by shaving and subsequent cleaning with antiseptics. It is possible to opt for either systemic or topical analgesia to alleviate any pain response due to the stimulus [105]. The animal is placed on the stage in an optimal position and immobilized ( Figure 4A). During laser-light treatment, both power (W/cm 2 ) and pulse interval (s) were manually adjusted in order to ensure target temperature with minimal fluctuations. As mentioned in the general considerations, a non-invasive method to determine the real-time hyperthermia profiles is MRT. After reaching steady state temperatures, the average tumor surface temperature can be monitored during laser light irradiation. In order to ensure MR compatibility, studies have designed holders, mounts and other features based on non-magnetic materials such as polyethylene and glass [41,100,106].
Similarly to CLS, mild thermal treatment via NIR laser light is mainly suitable for superficial tumors [107]. On one hand, the laser system is more advanced than CLS. Laser light is monochromatic and coherent, facilitating homogeneous illumination of the region of interest with distinct borders. However, on the other hand, the laser provides a highenergy light beam which, if not monitored properly, quickly leads to tissue dehydration and charring. This could in the long term lead to inhibition of energy delivery and side effects [106]. Therefore, it is essential to optimize the experimental settings, taking wavelength, beam size, and laser intensity into consideration. Table 3. Overview of the preclinical studies on application of local hyperthermia using NIR laser light sorted by year of publication and animal model. Similarly to CLS, mild thermal treatment via NIR laser light is mainly suitable for superficial tumors [107]. On one hand, the laser system is more advanced than CLS. Laser light is monochromatic and coherent, facilitating homogeneous illumination of the region of interest with distinct borders. However, on the other hand, the laser provides a high-energy light beam which, if not monitored properly, quickly leads to tissue dehydration and charring. This could in the long term lead to inhibition of energy delivery and side effects [106]. Therefore, it is essential to optimize the experimental settings, taking wavelength, beam size, and laser intensity into consideration.  3 if only measurements were provided, using the following formula: V = 0.5 (A·B 2 ). Estimates are indicated by "~".

Focused Ultrasound (FUS)
Focused ultrasound (FUS), which includes high-intensity focused ultrasound (HIFU), is a non-invasive technique based on local heat delivery to deep tumors via ultrasound waves [42,111] (Figure 5). Energy can be precisely deposited to the target area with minimal damage to intervening and surrounding soft tissues [42,112]. The only prerequisite is the presence of a fairly uniform medium (target tissue), as interference with either air or bone causes acoustic reflections or energy absorption. This could lead to non-uniform heat distribution and subsequent formation of undesired hot spots [42,78,113,114].
The FUS device consists of an imaging transducer for tumor identification and a therapeutic transducer for application of thermal treatment [115]. The ultrasound imaging probe can either be incorporated within the central opening of the FUS transducer or placed directly adjacent to it [116,117]. A distinction can be made between linear-and phased-array imaging transducers [118]. The linear transducer produces a rectangular beam shape, while the phased transducer is able to cover a larger area by either creating multiple focal points or steering a single, large focal zone in a specific direction [115,119]. For the therapeutic transducer, a distinction can be made between the single-and split-beam profiles. Transducers applying split beam technology are able to create a focal zone with a larger symmetrical volume (e.g., 1.49 mm 2 single focus vs. 7.4 mm 2 split focus), therefore directly reducing the number of ultrasound exposures required for treatment [42,120]. As demonstrated by Seip et al., application of asymmetric FUS geometries results in a lower maximum tissue temperature. However, this will, in turn, lead to a reduction in cooling time between subsequent sonications, resulting in treatment of a larger lesion volume in one treatment session ( Figure 5B) [121]. In order to minimize interference between the FUS beams and the ultrasound imaging, both transducers should not be operated simultaneously. US imaging is performed during the "OFF" period of the FUS pulse with an interval of~100 µs every 100 ms ( Figure 5C) [122].
Several factors should be taken into consideration for optimization of the insonification of the target area. First, depending on the set acoustic intensity FUS can be used for both ablation (high-intensity, >5 W/cm 2 ) and hyperthermia (low intensity, <3 W/cm 2 ) [42]. Second, the frequency (f) determines the depth of penetration. High-frequency pulses (1)(2)(3)(4)(5) tend to have a shallower penetration depth due to increased attenuation [123,124]. Third, the set duty cycle (DC) determines whether the ultrasound wave is in continuous or pulsed mode. Operating low duty cycles results in decreased temporal average intensities, and therefore generation of non-lethal temperature elevations [119,120]. Fourth, the applied thermal dose is of importance, which is dependent on applied maximum temperature and treatment duration. It should be taken into account that hyperthermia-mediated effects, such as enhanced extravasation, can last up to 24 h in all heating methods [125].
To realize the in vivo tumor model, the animal receives a subcutaneous tumor suspension or fragment, commonly either in the flank, mammary fat pad, or hind limb (Table 4). For instance, Chae et al. opted for tumor formation in the lateral part of the thigh to minimize the influence of breathing motions during FUS therapy [126]. Optimal ultrasound propagation from transducer to tumor occurs in the presence of a coupling medium. This should be a degassed liquid, either water, saline, or ultrasound gel, located in the gap between the FUS transducer and the target tissue [112,127,128]. The construction of a multilayered interface, which can be composed of ultrasound gel, agar gel, castor oil, mineral oil, or silicon oil, should even further decrease the intensity of backscattering [129]. Another possibility is the use of a coupling cone, which holds both the therapeutic transducer and the coupling medium [112,116,117]. In these systems, the depth of focus (5-50 mm) is determined by the configuration of the cone as both the distance between the transducer and the exit plane and the water pressure can be varied [112,116]. The pressure of the liquid surrounding the HIFU transducer determines the transmission speed of the ultrasound waves. As the transmitted waves have to travel through the coupling cone filled with degassed water, liquid pressure within the cone influences the depth of focus. It is optional to cover the aperture of the cone with a membrane film; however, coupling should still be ensured with the use of a coupling medium between the coupling cone and the tumor surface [112,117]. Another measure performed to improve acoustic coupling is removal of the animal's fur. A combination of shaving and application of depilatory cream prior to treatment minimizes interference of the ultrasound waves as it is susceptible to the presence of air [112,130].
The FUS therapeutic transducer transmits focused acoustic energy, which forms an ultrasound beam path through the coupling medium. Upon reaching the tumor surface, it produces a focal spot or treatment cell, which should cover the tumor and its margins [42,131]. The shape of the treatment cell is dependent on the geometry of the FUS transducer, frequently resembling a cigar-shaped ellipsoid (width: 1.5-2.0 mm; length: 1.5-2.0 cm) [42]. While a part of the energy will be either absorbed or reflected backwards, there is the possibility that the energy may penetrate deeper than the targeted area. To avoid this, a tumor may be divided into multiple treatment cells. Another option to prevent far-field heating due to acoustic reflections, or propagation into the contralateral thigh, is to opt for either a polyurethane rubber acoustic absorber or a saline bag to control the exit of the ultrasound wave. It is recommended to cover the absorber with a thin layer of isolating material to avoid indirect skin damage due to heat absorbance [114,[132][133][134]. Depending on the tumor location and the configuration of the HIFU system, the animal is placed on a platform in an optimal position and immobilized ( Figure 5A). Prior to therapeutic treatment, low-power test sonications can be performed in order to correct for focus point aberration [131,132,134,135]. Real-time hyperthermia profiles are determined with either an IR camera, thermocouples, or more commonly with MR thermometry (Table 4).

Capacitive Hyperthermia
Capacitive heating is based on generation of an electromagnetic field (8.00 or 13.56 MHz) applied by positioning the tumor-bearing part of the animal between two electrodes, with water or another substance at the tissue-electrode interface to guide the current through the animal. The amount of energy absorbed is dependent on electrical conductivity of the different tissues. It is postulated that absorption is higher in tumors, due to the Warburg effect-related increase in conductivity [140]. Besides by a standard continuous sinusoidal electromagnetic field, the energy can also be applied as modulated electro-hyperthermia (mEHT). mEHT is based on generation of an amplitude-modulated electromagnetic field, which is postulated to generate tumor-specific frequencies [40].
The capacitive setup consists of two plan-parallel electrodes. The lower electrode is frequently incorporated in the grounded, aluminum housing of the stage ( Figure 6A). Upon positioning the animal on the stage, there is indirect contact with the lower, static electrode. The stage is temperature-controlled, and connected to the capacitive system with heating and radiofrequency cables [141][142][143][144]. The upper electrode can either be a tissue-or pole electrode ( Figure 6B). A tissue electrode (TE) is a small, flexible circular element ( Figure 6C). It consists of a fabric cover ring, which encompasses a fabric with a conductive coating, such a copper-silver-tin [141,144]. The TEs are positioned over the tissue by rubber bands, which are stretched and subsequently anchored to the table [141,143,144]. As discussed by Danics et al. (2020), TEs could be inconvenient due to their large size and lack of adaptation to uneven surfaces [141]. Therefore, technical improvements have led to the design of a pole electrode (PE) ( Figure 6D). It consists of a column-shaped plastic casing, which houses multiple stainless-steel rods. At the tissue-electrode interface, the device is coated with a conductive fabric, such as silver-plated textile. In comparison to the tissue electrode, PE has a smaller conductive area and improved freedom of both placement and motility. This resulted in a more accurate tissue-electrode contact at curved areas, such as the inguinal region, which improves both focus and coupling [141].
In animal experiments involving capacitive systems, the animal receives a subcutaneous tumor either via suspension or fragment. Common tumor locations are the mammary fat pad, prostate, or hind limb (Table 5). To maximize coupling at the tissueelectrode interface and to minimize the induction of eddy currents, the treatment area benefits from shaving and the subsequent application of a liquid. This could be either ultrasound gel [143], conducting electrode cream [145], or a thin water bolus [146]. Another possibility is the placement of a thin solid insulator at the interface, such as cellophane [141]. Cooling of the upper electrode could be realized using dampened gauze on the TE [143] or incorporation of Peltier elements in the PE [146]. Unlike other hyperthermia devices, external shielding is not required to prevent dispersion as the current flows through a closed, directional electric circuit [147].
Capacitive devices can be applied to various anatomical sites to target both superficial and deep-seated tumors, depending on the diameter of the opposed electrodes [148]. However, the amount of subcutaneous adipose tissue at the target site should be taken into account. As the electric field has to pass through this high-resistance layer, the absorbed energy causes substantial heating [73,149,150]. Upon establishment of the electric field, a current will flow through the target area from one electrode to the other. Being a closed, directional circuit, the formation of undesired hot spots outside the electric flow is limited [151]. Upon passing through the dielectric media (tissue), there will be differences in the permittivity and conductivity [40]. The electromagnetic current will pass through surfaces with a low impedance (high conductivity) as it will by default choose the path of least resistance.
Other types of electromagnetic heating, such as radiofrequency and microwave, have also been employed for preclinical hyperthermia studies [37,152,153]. Radiofrequencybased hyperthermia (3 Hz to 300 MHz) is most frequently used for deep-seated tumors, while microwave heating (300 MHz to 300 GHz) is suitable for treatment of superficial malignancies [154]. Both types of electromagnetic heating may be used for advanced drug delivery studies in combination with intravital microscopy or MRI for real-time visualization. These techniques are discussed in detail by Priester et al., see elsewhere in this special issue.      3 if only measurements were provided, using the following formula: V = 0.5 (A·B 2 ). Estimates are indicated by "~".

Conclusions
The standardization of mild hyperthermia represents further improvement of preclinical testing, as some attribute disappointing clinical trial results to the lack of supporting preclinical data [158]. This review provides a preclinically oriented detailed overview on systems for solid tumor hyperthermia treatment in small animals (Table 6). By discussing only original research articles, which present a detailed description, a new design, or an essential update on an existing method, standardization is facilitated by addressing both general considerations and method-specific details.
The devices described in this review can be ranked based on technological complexity. Whereas water bath heating is considered to be a low-complexity, entry level device, it is an accurate system for homogeneous yet non-specific heating. Another advantage of this technique is the possibility to simultaneously treat multiple animals, thereby minimizing variation due to exposure to the same conditions [58]. Therefore, the water bath method is suitable to efficiently determine the effects of localized mild hyperthermia. However, it is only a small step progressing from a water bath to a CLS device with distinct advantages. The simple CLS option in the form of a halogen-based device can easily be obtained by making small alterations to an existing light source. The inclusion of a water filter may even further optimize the device, but both are an improvement considering penetration depth, target specificity, and device tuning.
The use of a NIR laser-based device even further enhances HT application, due to an increased penetration depth, a more precise focal zone, and the tuneability of multiple factors. However, thorough optimization of the device is necessary in order to avoid side-effects such as irreversible skin damages. FUS is the most advanced technique in this group, and the versatility allows for thermal treatment of both superficial-and deep tumors. Comparable to the NIR laser-based system, optimization of several settings is required. However, in the end, FUS is a more advanced system with favorable penetration depth, high target specificity, and the possibility to include either a diagnostic transducer or MR acquisition for spatial control and thermometry. However, preclinical studies on FUS are mainly performed in large animals ( Figure 1). Similar to FUS, capacitive hyperthermia is a non-invasive technique, which can be used for thermal treatment of both superficial and deep-seated tumors. Capacitive heating takes advantage of tissue dielectric properties, resulting in (non)-thermal therapeutic effects at the target site. However, it should be taken into account that adipose tissue, being a thermal insulator, should be circumvented to avoid substantial heating.
Taken together, the described basic hyperthermia devices provide an excellent foundation for preclinical research when carefully considering the benefits and limitations. The advantage of more advanced devices for certain applications may be worthwhile, however this advancement obviously results in a more demanding, complicated, and costly setup.