LEPABE, CI-IPOP, University of Porto, Portugal
INEGI, University of Porto, Portugal
CITAB, QOPNA University of Trás-os-Montes and Alto Douro, Portugal
CITAB, University of Trás-os-Montes and Alto Douro, Portugal
INEGI, University of Porto, Portugal
Biophysics Institute, University of Coimbra, Portugal
CIMAGO, CNC.IBILI, University of Coimbra, Portugal
Laboratory animals provide important models for studying human diseases, including many types of cancer. Mice are among the most commonly used laboratory animals, allowing for the study of carcinogenic agents, cancer development and for testing innovative preventive and therapeutic strategies. Thus, monitoring angiogenesis in animal models is a major goal for cancer research. Among the currently available imaging techniques, thermography is a useful approach for studying the superficial vascularization of cancer, based on their heat emissions. At this chapter emphasis is placed on thermography and its applications on laboratory animals, in comparison with other available and applicable imaging techniques. In conclusion, thermography may be usefully applied to the study of cancer vascularization in animal models, particularly when using laboratory rodents such as mice. Care is needed in adapting existing approaches to the specificities of each animal species.
Among the major public health problems of our time, cancer occupies a prominent place. Cancer develops through a complex, multistage process, driven by the cellular accumulation of genetic mutations and epigenetic events (Bignold, 2003; Bignold, Coghlan, Jersmann 2006; Oliveira et al., 2007; Iacobuzio-Donahue, 2009). This process of carcinogenesis often proceeds through multiple distinguishable morphological stages as shown in Figure 1. Thus, pre-cancerous lesions may develop into benign and later malignant tumors. The term cancer is reserved for malignant lesions, while the term neoplasia (used of a tissue which has become independent of normal physiological regulatory mechanisms) applies to both benign and malignant tumors. Hence, early non-tumoral lesions are often referred to as pre-neoplastic. Environmental carcinogens are biological (e.g. human papillomavirus, HPV), chemical (e.g. nitrosamines) or physical (e.g. ionizing radiations) agents that initiate carcinogenesis by inducing genetic mutations or promote it otherwise (Oliveira et al., 2007). Genetic mutations occurring in healthy cells are, most often, successfully repaired or else they trigger programmed cell death mechanisms (as a way to prevent carcinogenesis). However, unrepaired mutations that block DNA repair mechanisms (encoded by so-called tumor suppressor genes) will promote the unchecked accumulation of further mutations (genomic instability). Mutations that activate genes (so-called proto-oncogenes) that contribute to make the cell independent from its environment (e.g. by producing its own growth factors) are also important in driving carcinogenesis. Accordingly, during the early phases of carcinogenesis, cells tend to survive and proliferate in an unregulated fashion, accumulating additional mutations (Oliveira et al., 2007) that will drive carcinogenesis further on Figure 1. As already pointed out, neoplastic tissues are not necessarily malignant. While benign tumors are well-delimited and damage adjacent tissues mainly by compressing them, malignant tumors (cancers) are able to invade adjacent tissues or event to spread to distant body parts through lymphatic or blood vessels. Progression from a benign to a malignant stage is commonly observed in many epithelial tumors. The malignant phenotype requires an ability to interact with and invade the adjacent connective tissue (called the stroma). When neoplastic cells have acquired the typical traits of malignancy, they degrade the basement membrane that separates them from the stroma, and spread into the adjacent tissues. By invading blood and lymphatic vessels (a phenomenon called intravasation) in the stroma, malignant cells may gain access to distant organs, where they may establish distant tumor foci, called metastasis. The interaction between malignant or pre-malignant cells and the adjacent stroma is highly complex.
In order to study these phenomena, researchers have long resorted to employing animal models, which allow for fast and efficient experimental approaches. Thermography, as an imaging technique, is able to provide valuable additional information on the physiology and pathology of these models, namely concerning superficial inflammation and blood irrigation. The present chapter deals with laboratory animal models of cancer, followed by multiple imaging techniques available for use in these models. Special emphasis is given to the use of thermographic techniques to evaluate laboratory mice and rats.
One key feature of the interaction between cancer cells and their adjacent stroma is the development of new, and often abundant, blood vessels, termed angiogenesis (Potente, Gerhardt, Carmeliet, 2011). Tumor-associated angiogenesis is of particular importance clinically and also from the viewpoint of thermographic analysis, as it provides the basis for temperature differences between healthy and tumoral tissues (Carmeliet & Jain, 2011).
|Figure 1. An overview of multistage cancer development in epithelial tissues. a- normal stratified epithelium (blue), with basal cells covered by differentiated cells. The epithelium lies over a basement membrane, which separates it from the underlying stroma, containing fibroblasts (yellow) and capillaries with red blood cells. b- one epithelial cell suffers DNA mutations, triggering carcinogenesis. c- loss of normal epithelial structure, and development of multiple cellular clones harboring additional different mutations. d- an aggressive, angiogenic cellular clone develops, secreting pro-angiogenic factors and re-structuring the adjacent stroma, which shows numerous capillaries, fibroblasts and leukocytes. e- malignant (cancer) cells degrade the basement membrane, invading the stroma and blood vessels. d- cancer cells migrate through the blood and lymph vessels, establishing metastasis in distant organs|
Angiogenesis is essential to deliver nutrients to tumor cells. Tumors do not grow beyond 1 or 2 mm in diameter without developing their own blood supply. Beyond that size, oxygen and nutrients do not diffuse from adjacent blood vessels in the necessary amount, resulting in hypoxia and cell death (Potente, Gerhardt, Carmeliet, 2011). In fact, many tumors are hypoxic and necrotic areas where the blood supply is limited. A number of pro-angiogenic and anti-angiogenic factors exist, that govern the development of arrest of angiogenesis in physiological and pathological conditions. During tumor development, some tumor cells acquire the possibility of modifying the balance between pro-angiogenic and anti-angiogenic factors in favor of angiogenesis, a phenomenon known as the angiogenic switch (Shojaei, 2012). In order to achieve this, tumor cells (and, frequently, certain stromal cells) secrete a number of pro-angiogenic factors, most importantly vascular endothelial growth factor (VEGF) and basic fibroblast growth factor (bFGF). Down-regulating the expression of anti-angiogenic factors is also an important strategy, e.g. mutational inactivation of the p53 protein leads to decreased production of the anti-angiogenic protein thrombospondin-1 (Grossfeld, 1997). VEGF acts through its receptor (VEGFR) to attract circulating endothelial precursor cells, drive the proliferation of endothelial cells (which line the lumen of blood vessels) and their differentiation at the tumor site, as well as promoting the sprouting of new capillaries from existing blood vessels (Potente, Gerhardt, Carmeliet, 2011). The blood vessels that develop show an irregular, tortuous, abnormal architecture and increased leakiness, compared with blood vessels from healthy tissues (Shojaei, 2012). In some cases, tumor cells may line structures resembling capillaries, a phenomenon known as vasculogenesis mimicry. Besides delivering oxygen and nutrients to tumor cells, endothelial cells also provide them with important polypeptide growth factors, such as insulin-like growth factor and platelet-derived growth factor. Furthermore, the development of an extensive and accessible capillary bed facilitates tumor intravasation and metastasis.
According to the most recent GLOBOCAN data, cancer incidence rates are consistently increasing, both in developed and in developing countries, with an estimated 14.1 million new cancer cases diagnosed in 2012 (Torre, 2015). The leading causes of death by cancer are lung, breast and colorectal cancers. The rising cancer incidence is believed to reflect the increasing life expectancy and changing life habits, especially the adoption of cancer-causing behaviors such as smoking. Cancer mortality tends to be higher in developing countries, due to limited access to early diagnosis and to adequate therapy (Torre, 2015). Overall, during 2012, cancer caused an estimated 8.2 million deaths worldwide (Torre, 2015). Breast cancer is now the leading cause of death by cancer among women in developed countries. Screening and vaccination programs are good examples of how understanding the biopathology of cancer were translated into effective preventive strategies. Another highly significant example, is the development of anti-angiogenic therapies, mainly based on blocking VEGF-mediated signaling (Carmeliet & Jain, 2011). The main challenges in this area are inherent or acquired resistance to angiogenesis inhibitors, a possibly increased tumor aggressiveness during anti-angiogenic treatment, and the absence of validated biomarkers to select suitable patients and monitor tumor responses (Shojaei, 2012).
Animal Models of Cancer: General Concepts
In vivo animal models of cancer are critical tools to study the biopathology of cancer and the effects of carcinogens, as well as for developing innovative preventive and therapeutic anti-cancer strategies. In the last years, with the development of genetics, sophisticated animal models have been developed that better represent specific cancer features. Nowadays, animal models of cancer come in a diversity of forms, ranging between cancer cell lines and genetically engineered animals.
Animal models provide an alternative mean to understand the causes and to evaluate new treatments for cancer, being a resource of a great potential in oncology. The use of animal cancer models has provided a better understand of cancer’s biology and, more recently, its genetics (Frese & Tuveson, 2007). Several animal species, namely dogs, large animals, rabbits, fish, guinea pigs, hamsters, rats and mice have been used in experimental protocols, the rat and mouse being most frequently used (Decker & Sausville, 2011). Rats and mice have some advantages when compared with other animal species. They are mammals, small, their accommodation and maintenance are cheap, they are easy to handle and reproduce, have a high number of offspring, their physiology and genetics are well understood and are similar to humans in various aspects, namely concerning their anatomy, physiology, genetics and biochemistry. The similarities between such animal models and humans allow their use to find new diagnostic methods, as well as discovering and testing new therapies (Oliveira, Colaco, De la Cruz, Lopes, 2006).
A wide variety of animal models of cancer is currently available. In vivo cancer animal models can be divided into different categories: genetically engineered, transplantation (syngeneic and xenotransplant), spontaneous and carcinogen-induced models (Table 1). Researchers may select the model according their investigation’s aim.
Table 1. Characteristics of animal models of cancers: some advantages and disadvantages
|Genetically engineered||Tumor arise “naturally”.|
Tumor and stromal cells interact in their normal microenvironment.
Closely mimics human cancer pathogenesis.
|Mutation usually only in a single gene, failing to mimic the genetic complexity and heterogeneity of human tumors.|
Tumors arise at different stages.
|Transplantation||Syngeneic||Tumor cells and the host share a common genetic background.|
Host immune system is not compromised (immune-competent host)
Mimics the normal tumor microenvironment.
Rapid development of tumors.
|Tumor cells are non-human|
In some cases, tumors are implanted in a non-natural site.
|Xenograft||Tumor cells are from human.|
Wide variety of tumor cell lines available.
|Host is an immune-deficient animal.|
Longer latency period (compared with syngeneic model).
In some cases, tumors are implanted in a non-natural site.
Does not mimic normal tumor microenvironment.
Lack of protective immune response and tumor-promoting inflammation.
|Spontaneous||Allow the identification of cancer susceptibility genes.||Long latency period.|
Difficult to obtain a sample with an adequate size.
|Carcinogen-induced||Short period of latency.|
Allows the study of carcinogenesis (e.g. mutagenesis, oxidative stress, inflammation).
Allow the identification of potential targets for cancer therapy.
In some models, induced tumors are internal and difficult to monitor.
Some models simultaneously develop tumors in multiple organs.
Following the identification of many mutations responsible for the development of some human cancers, transgenic and gene targeting technology allowed the development of a large number of genetically engineered rodent strains. These genetically engineered cancer models express oncogenes or mutations in tumor suppressors found in human tumors (Hanahan, Wagner, Palmiter, 2007). In these genetically engineered animal models there is the interaction between tumor and stromal cells in their normal microenvironment, which mimics the human cancer pathogenesis better than in transplanted models. However, genetically engineered rodent strains have some limitations. Since these models usually have mutations in a single gene, they do not accurately mimic the genetic complexity and heterogeneity of human tumors. In order to overcome this limitation, mouse strains possessing combinations of oncogenic mutations have recently been developed. However, these genetically engineered animal models are expensive and not yet affordable for many research teams (Farago, Snyder, Jacks, 2012; Singh, Murriel, Johnson, 2012).
In transplantation models, tumor cell lines or a part of a human tumor, may be transplanted into immune-deficient (xenograft models) or immunocompetent (syngeneic models) animals. When tumor cells are well accepted, they result in tumor formation within a few weeks. According to the aim of the researchers, there are several ways to transplant cells: subcutaneously (heterotopically) to allow the study of tumor development, intravenously to mimic the formation of metastasis, or orthotopically when the tumor cells are transplanted to their organ of origin, mimicking tumor development in its own microenvironment (Loi et al., 2011).
In xenograft models, human tumor cells are transplanted into an immune-deficient animal. These models do not mimic the natural tumor microenvironment due to a lack of protective immune response and tumor-promoting inflammation (Dranoff, 2012).
In syngeneic models, the tumor cells and the host share an inbred genetic background. The host immune system is not compromised and transplanted tissues are not rejected by the immune system of the host. This model mimics the normal tumor microenvironment. These models are rapid, reproducible and affordable when compared with xenografts and genetically engineered models (Polin, 2011).
Decades of studies in animal models allowed researchers to discover that some animal species and strains develop cancer in a certain organ or in various organs when exposed to a specific agent, under appropriate conditions, namely age of the exposure, number of exposures/administrations, dose or time of exposure. The agents that have the capacity to induce the formation of tumors are classified as carcinogen agents. These agents can be chemicals (e.g. polycyclic aromatic hydrocarbons, nitrosoureas, nitrosamines, cadmium, arsenic, tobacco compounds), virus (e.g. mouse mammary tumor virus), radiations (e.g. ultraviolet radiation, ionizing radiation) or minerals (e.g. asbestos fibers). The tumors in carcinogen-induced models have a short period of latency and are easily reproducible (Jonkers & Berns, 2005).
Spontaneous cancer animal models are animals that naturally develop cancers, these models allow researchers to identify the genes that generate susceptibility to cancer development (Hansen & Khanna, 2004). Spontaneous tumors in dogs and cats are frequently used in cancer research, as these animals frequently develop tumors with histopathologic and biologic behavior similar to those that occurs in humans (MacEwen, 1990).
Despite the limitations of each type, in vivo animal models of cancer are invaluable tools to study the biopathology of cancer, understand the complexity of carcinogenesis, the effects of different carcinogens and to develop innovative preventive and therapeutic anti-cancer strategies.
Imaging Techniques in Animal Models of Cancer
Medical imaging allows the creation of visual representations of the anatomy and physiology of the interior of a body. The technology, which may or may not apply ionizing radiation, is used as diagnosis tool to identify abnormalities in comparison with normal conditions. Besides this application, it is also possible to perform imaging of removed organs and tissues using pathological techniques.
The discovery of different types of radiation and its use as a diagnostic and therapeutic methods constituted a significant advance in medicine. Small animal-dedicated scanners were produced only in the late 1990s. However, the large number of research centers interested, quickly made this equipment very popular. Therefore, nowadays there are dedicated animal cameras for all the main imaging techniques, such as microSPECT, microPET, microMRI, microCT and thermography.
The X-rays are electromagnetic radiation similar to visible light, but with a shorter wavelength. Considering its particular characteristics, this type of radiation allows both, to obtain images, and also to be used for therapy (radiotherapy). Thus, its use as a diagnostic tool is based on the ability of the X-radiation to propagate throughout the body. The penetrating power of X-ray is inversely proportional to its wavelength, and when a volume is exposed, some portion of the radiation energy is absorbed, while the other is scattered through the matter (Lima, 1995).
The obtained image is defined as conventional radiography, and is achieved after the X-rays pass through the object and the emerging photons are able to cause fluorescence in enhancing screens. The emitted light strikes the emulsion on a film, giving rise to a latent image. Computed radiography differs from this conventional technique because the emerging photons interact with a phosphor plate, thus creating the latent image which is automatically converted into a digital image through a scanner (Gaivão, Lima, Agostinho, Oliveira, Peres, 2001).
Concerning animal applications, the X-ray is one of the most commonly used diagnostic tools, with the possibility of obtaining 2-dimensional images which are representations of 3-dimensional objects. Due to its characteristics, this technique provides a large amount of information through non-invasive and affordable means. Besides this, X-ray do not change the disease process and does not cause pain. The results obtained with X-rays imaging depend on the combination of different parameters such as the beam voltage, the number of X-rays produced and the exposure time. Besides these instrumentation characteristics, the tissue density also influences the obtained image (Lima, 1995; Gaivão, Lima, Agostinho, Oliveira, Peres, 2001).
To perform animal imaging using X-ray (Figure 2), the animal must be adequately restrained and positioned to obtain high-quality radiographic images, whereby sedation or short-acting anesthesia is often necessary. Besides that, image acquisition parameters may also be manipulated in order to minimize the effects of animal motion. Specifically, these conditions are the decrease of exposure time and the maximization of the power (Merck Sharp, Dohme Corp, 2014).
|Figure 2. Skull radiograph of the mouse in profile with changes in bone structure suggesting bone involvement by an orthotopic osteosarcoma model|
Computed Tomography (CT)
Since the 1970s, CT has been established as an imaging approach which produces axial slice images using a source of X-rays. Using this imaging technique it is possible to estimate the linear attenuation coefficient of the X-rays in each pixel of slice, which are obtained around the patient considering different angles. Whereas the final images on radiographs are obtained by the compression of a volume in a 2D level, in which the radiological projection information from a three-dimensional object is represented, with CT it is possible to get a flat image of a three-dimensional object visualized in a gray scale. This imaging methodology, allows a better temporal and spatial resolution, more anatomic imaging, improved signal/noise ratio as well as a better image processing potential and a retrospective reconstruction, especially taking into account the innovations using multi-slice imaging and newly developed contrast media (Gaivão, Lima, Agostinho, Oliveira, Peres, 2001).
For animal imaging studies, all CT scanners that provide higher spatial resolution than current scanners used in clinics are named micro-CT. The spatial resolution of these equipment is in the order of a few hundred µm. In this imaging technique, anesthesia is not required for intravenous, intraperitoneal or oral administration of contrast agents. However, mice or rats are imaged under anesthesia. To perform a CT study, animals are positioned within the scanner, and visually monitored during the scan (Bartling, Stiller, Semmler, Kiessling, 2007). Employing this imaging technique, it is possible to design in vivo research experiments allowing repeated measurements of the same animal. This approach reduces the variability associated with images coming from different animals (Bartling, Stiller, Semmler, Kiessling, 2007).
Nuclear Medicine (NM)
Nuclear medicine it is a diagnostic imaging and therapeutic technique also known as molecular medicine or molecular imaging, which allows in some situations, the performance of metabolic radiotherapy. This technique uses properties of radioactive isotopes and the energetic radiation emitted by them. This is always ionizing radiation and may be of different natures (particles or electromagnetic). Nuclear medicine uses both types of radiation for different purposes, the particles for therapy and the electromagnetic (gamma rays and annihilation photons) for imaging. Concerning the utilization of X-rays, radiology is more dedicated to morphological studies, whereas nuclear medicine enables the assessment of functional information. Nuclear medicine can be divided into single-photon emission computed tomography (SPECT) and positron emission tomography (PET), each technique associated with a specific and/or dedicated equipment. In nuclear medicine using SPECT, images are acquired after collimation of the photons, which are detected by a crystal that gives off a light signal. This is, in turn, amplified and converted into count data, planar or, after processing, 3D tomographic images. This imaging technique allied to a CT scanner can provide information about localization and function simultaneously. In SPECT imaging, Technetium-99m is the radionuclide most used. This isotope labels the majority of the molecules with different biodistributions (Lima, 2011) (Figure 3).
|Figure 3. Nuclear medicine (SPECT) imaging using 99mTc-MIBI. Static image of an animal model with colostomy and colorectal cancer with 99mTc- MIBI uptake in tumor localization|
PET uses a short-lived positron-emitting isotope such as Fluor-18, which is the most commonly used (Figure 4). Modern scanners used in small animal imaging may integrate PET with CT and MRI, allowing the production of PET-CT or PET-MRI images. In small animal imaging equipment, these combinations are possible and can all be simultaneously present in the same equipment without physically moving the animal off the gantry. This possibility allows the generation of images that match information from different sources. The resultant hybrid image with functional and anatomical information is a useful tool in non-invasive diagnosis and provides information over time, promoting respect for the 3Rs principles of animal research ethics (Lima, 2011; Rowland, Cherry, 2008, May).
|Figure 4. Nuclear medicine (PET) imaging using 18F-FDG. Static image of an animal model with colostomy and colorectal cancer with 18F-FDG uptake in tumor localization, 60 minutes after radiopharmaceutical administration|
Since the discovery of the piezoelectric properties of quartz in 1880 by Marie Curie until its current status, ultrasound imaging is the result of several decades of research. This medical imaging uses high frequency sound waves in the megahertz range that, once applied on tissues, are reflected to various degrees, being useful to produce 3D images (Gaivão, Lima, Agostinho, Oliveira, Peres, 2001).
Employing the transmission and reception of sound pressure waves it is possible to perform tissue characterization, using frequencies ranging from 20 kHz up to several GHz, which allow for a different medical image modality. The high frequency sound waves are applied on tissues and, depending on the composition of the different tissues crossed, the signal is differently attenuated, and the returned echo is also separated by different intervals. This imaging approach is safe, not appear to cause any adverse effects in the body. Besides that, it is also relatively inexpensive and quick to perform and the equipment does not use ionizing radiation (Gaivão, Lima, Agostinho, Oliveira, Peres, 2001). However, the final quality of the image depends on the technician skills.
Concerning ultrasonography for animals, higher frequencies are used when compared with clinical devices used in human patients. Besides that, there is the possibility to use different contrast agents in order to study specific pathologies related with cellular receptors. This imaging approach has also the advantage of enabling real-time imaging, with the possibility of capturing about 1000 frames per second and allowing the visualization of blood flow in vivo, which is adequate, for instance, for cardiac studies. Considering its spatial resolution, ultrasonography is also suitable to study angiogenesis related with tumor development. However, the limited depth of penetration, influences the type of pathology in which ultrasound can be used, and should be considered as an inconvenient characteristic of this technique (Gaivão, Lima, Agostinho, Oliveira, Peres, 2001; Renault et al., 2006).
Magnetic Resonance Imaging (MRI)
A magnetic resonance imaging instrument (MRI scanner), or a nuclear magnetic resonance (NMR) imaging scanner as it was originally known, uses powerful magnets to polarize and excite hydrogen nuclei (single proton) in water molecules of biological tissues, producing a detectable signal which is spatially encoded, resulting in images of the body. The MRI machine emits a radio frequency (RF) pulse that specifically tilts the magnetic moment of the hydrogen nuclei. This pulse causes the protons in that area to absorb the energy needed to make them spin in a different direction. This is the “resonance” part of MRI. The resonance frequency is called the Larmor frequency and is calculated based on the particular tissue being imaged and the strength of the main magnetic field. MRI uses three kinds of electromagnetic fields: a very strong (in the range of a few tesla) static magnetic field to polarize the hydrogen nuclei, called the static field; one or more weaker time-varying (on the order of 1 kHz) field(s) for spatial encoding, called the gradient field(s); and a weak radio-frequency (RF) field for the manipulation of the hydrogen nuclei to produce measurable signals, collected through a RF antenna (Gaivão, Lima, Agostinho, Oliveira, Peres, 2001).
As for CT imaging, MRI traditionally creates a two dimensional image of a thin “slice” of the body and is therefore considered a tomographic imaging technique. Modern MRI instruments are able to produce images in the form of 3D blocks, which may be considered a generalization of the single-slice, maintaining the tomographic concept. However, unlike CT, MRI does not involve the use of ionizing radiation and is therefore not associated with the same health hazards. On the other hand, because MRI has only been in use since the early 1980s, there is not much confidence what are the effects of long-term exposure to strong static fields. Because of that, there is currently no limit to the number of scans to which an individual can be subjected, in contrast to what happens with radiology, CT and nuclear medicine. However, the health risks associated with tissue heating due to exposure to the RF field and to the magnetic field, are well-identified, namely if there are devices implanted in the body, such as pacemakers or other metal apparatus. These risks are strictly controlled as part of the design of the instrument and the scanning protocols used (Gaivão, Lima, Agostinho, Oliveira, Peres, 2001; Chatham, Blackband, 2001).
Because CT and MRI are sensitive to different tissue properties, the appearance of the images obtained with the two techniques differ markedly. In CT, the X-rays are more attenuated by dense tissue to create an image, whereby the image quality of soft tissues is poor. In MRI, although any nucleus with a net nuclear spin can be used, the proton of the hydrogen atom remains the most widely used, especially in the clinical setting, because it is ubiquitous and returns a large signal. This nucleus, present in water molecules, allows the excellent soft-tissue contrast achievable with MRI (Gaivão, Lima, Agostinho, Oliveira, Peres, 2001; Bartling, Stiller, Semmler, Kiessling, 2007).
Pathological analysis is older than any of the imaging techniques already described. Using a tissue sample, it is possible to perform a morphological evaluation of tissues in search for clues concerning its histogenesis and biological behavior. Nowadays, the pathology is able not only to label diseases but also to characterize and define them morphologically as well as elucidates their underlying mechanisms (Cross, 2013; Van Middendorp, Sanchez, Burridge, 2010).
When using body tissues for pathological analysis, it is necessary first to fixate them before proceeding to histological analysis through microscopy. Fixation allows the preservation of tissues permanently in a life-like state.
Over the years, the role of pathology in medicine has benefitted from the application of specialized techniques allowing a more complete characterization of the specimen. Both, enzyme histochemistry and electron microscopy, have expanded the role of pathology from the primary microanatomical (histological) evaluation to the assessment of biochemical and subcellular structural features, exploring immunological markers and chemical signatures of cells. Over the years, advances on fixation, embedding, cutting, immunohistochemical staining, molecular methods, microscopy, as well as in image processing have improved the diagnostic ability and the research applicability of pathology (Cross, 2013; Van Middendorp, Sanchez, Burridge, 2010).
In animal research, pathological approaches have an add value, in order to obtain a complete tissue characterization in comparison with the other imaging techniques available (Figure 5). However, it is necessary to keep in mind the implications of performing a technique which involves a highly invasive procedure, i.e. removing a sample of tissue. Accordingly, pathological analysis is often performed at the end of experimental protocols or when animals are sacrificed at intermediate time points.
|Figure 5. Representative images of histology specimens obtained from an orthotopic colorectal cancer animal model, stained with hematoxylin and eosin (H&E) and isolated from: colorectal cancer (A), liver (B), lung (C) and lymph node (D). Magnifications: 40x|
The image techniques seen before can be processed in order to get some useful medical information that can support the diagnosis:
- • Geometry,
- • Density,
- • Flow,
- • Size,
- • Symmetry,
- • Of the organ or zone of interest.
Thermography is an imaging technique that captures infrared radiation and uses it to create a temperature mapping image. Therefore, thermography not only reads the thermal patterns of a certain zone but simultaneously quantifies the temperature in each of the image picture. This adds value to the technology but requires correct equipment, conditions and protocols to perform a rigorous and useful thermographic analysis.
The majority of published thermographic studies tries to correlate a temperature variation (increase or decrease) with a certain pathology or physical disorder (Snekhalatha, Anburajan, Venkatraman, Menaka, 2013). In fact, since thermography measures radiation to create an image representing the temperature of objects, it is the most suitable method to assess the thermogenesis and thermoregulation in the majority of animals. However, the thermographic measurement of the body temperature is a challenging task as it depends on the environment, animal fur, and a precise protocol.
Infrared Measuring Principle
Temperature is due to the kinetic movement of the molecules inside the objects. As this movement increases in amplitude and frequency, the object’s temperature also increases, and thus the infrared radiation emission. The correlation between the temperature and the spectral radiation distribution can be described by the Plank law. This law states that the amount of radiation and the distribution peak frequency increases proportionally with the temperature. This way, it is possible to calculate the temperature of an object by measuring the amount of radiation in a specific waveband. A thermal camera captures infrared radiation and converts it into an electric signal that is then converted into a digital value. Through a calibration procedure, using a black body, it is possible to calculate the temperature at every pixel of the thermal image. A black body is an ideal physical object that absorbs all incident radiation, regardless of the incidence angle or radiation frequency. As an output, the emitted radiation follows the Plank law perfectly and is only modified by changing the black body’s temperature (Figure 6). Another important aspect is that the radiation density is not dependent on the viewing angle. Even if the radiation is not isotropically distributed, this aspect is usually ignored since the camera’s viewing angle is relatively small (less than 30º) and therefore their inherent errors are acceptable. The emissivity can then be seen has the “closeness” that an object is from being a black body, resulting in values from 1 (black body) to 0.
|Figure 6. Black body radiation distribution at several temperatures, according to the Plank’s law. LWIR – Long Wavelength Infrared (8-14 μm), represents the main waveband used in quantitative infrared thermography as it corresponds to the peak for environment temperatures|
The majority of the studies take place at room temperature (approximately between 293 and 297 K). Therefore, it is necessary to use a camera that is able to capture the maximum amount of radiation in that temperature range, in order to obtain readings as precise as possible. According with the Plank law represented in Figure 6, the majority of the emitted radiation, at the ambient temperature (approximately 300K) is located between the 7 and 17 µm of wavelength. This type of radiation is known as infrared radiation (ranging from 0.77 to 1000 µm), (Ryer & Light, 1997).
Since thermography uses and measures radiation to calculate temperature, it is influenced by some properties of the object, the environment and the camera itself. The main factors influencing thermographic measurements are:
- • Objet
- o Emissivity
- o Reflected temperature and radiation
- o Emissivity
- • Environment
- o Environmental temperature
- o Relative humidity
- o Distance
- o Environmental temperature
The emissivity is the balance between the emitted radiation due to the object’s temperature and the radiation reflected and transmitted through the object. The skin emissivity is very high, approximately 0.98, and is the most important parameter to be set when acquiring thermal images. Another important aspect is the reflected radiation, emitted by any object in the vicinity, which interferes with temperature measurements. This effect also takes place due to background radiation emitted by the room walls, especially if the room temperature is far from the recommended. Depending on the study, and mainly the animal being studied, the ideal room temperature may vary. Although the room temperature is generally assumed to be stable during the entire image acquisition process, the reflected radiation often changes (e.g., when the hands of the researcher manipulate the animals). This aspect is far more important when working with mice compared with human patients. As mice are so small, researcher’s hand occupies a considerable part of the image, thereby increasing the errors and uncertainties associated with the measurements. Ideally, animals should be pictured without any handling, which often requires sedation or a light transient anesthesia (e.g. volatile anesthesia using isoflurane). Care is needed to avoid prolonged anesthesia protocols, as these interfere with body temperature regulation.
Ideally, the lenses of the thermal camera should be completely transparent to the infrared radiation. In reality, this is impossible and the lens’ transmissibility should be considered in temperature calculations. Consequentially, part of the radiation received by the sensor will derive from the lenses and not from the object. This is the reason why the thermal camera should be kept at a stable temperature.
Thermography Operating Issues
One of the most important procedures when acquiring a thermal image is the focus. An unfocused image can lead to increased uncertainty or even wrong evaluation of the thermal image. If the temperature is important for interpretation, this aspect is even more important. Figure 7, shows two thermographic images of a Wistar rat presenting chemically-induced mammary tumors in an early phase. The animal was partially shaved over the tumor area, however hair scatters the emitted infrared radiation and severely affects image quality. In Figure 7a) it is possible to observe the shaved area exhibiting three hot areas and a central colder zone. However, after adjusting the focus to the central colder area, it shows up as a clump of unshaved hair, interfering with image analysis (Figure 7b). Shaving the animal may cause light trauma to the skin, with associated inflammation, thus increasing local temperature. Care should be taken to allow sufficient time for the inflammatory phenomena to abate (usually up to 48 hours should be enough).
|Figure 7. N-methyl-N-nitrosourea-induced model of mammary cancer in a female Wistar rat. a) poorly focused thermal image of the tumor, b) correctly focused of the central area of the tumor|
Another important aspect is the correct distance from the object. To achieve a perfect focus, the area to analyze is expected to be as flat as possible, which is more relevant as the camera approaches the animal. However, due to the natural round form of rats and other small animals, this situation is not usually verified. In order to minimize this problem, the object to be analyzed should be positioned as far as possible, without losing the necessary detail. An example of this situation is represented in Figure 8, representing a mouse skin tumor. The lesion was imaged using a macro lens in order to emphasize the referred effect: by focusing on the tumor basis and contour, surface and center areas become blurred. The ideal situation would be to place the camera at a 90º angle from the animal’s surface, which is difficult, in this case, due to the natural shape of these animals.
|Figure 8. Two-stage skin cancer model developed with 12-O-tetradecanoylphorbol-13-acetate and 7,12-dimethylbenz(a)anthracene in a mouse. a) Poorly focused image of the tumor, b) Correctly focused image of the tumor|
Image Acquisition Modalities
Thermal images may be acquired for several purposes, therefore the acquisition methodology and modality should be chosen for each case. Thermal imaging modalities can be divided into passive and active thermography (Figure 9).
|Figure 9. Thermography modalities|
In passive thermography, a single thermal image is enough to detect and characterize the object of analysis. Skin temperature changes may be detected in consequence of anomalies in the skin or in organs near the surface. For this reason, passive thermography has been the most used method in health applications, being also the easiest method to use.
On the other hand, active thermography uses a heat flow to induce temperature changes and thus increase the accuracy of thermal readings, or to study a particular aspect e.g. thermoregulation phenomena.
Active thermography can be divided in two modalities, singular and cyclic stimulation. In the first case, the object is submitted to a single thermal stimulation; its reaction and its recovery after stimulation are then observed. Thermal transient analysis is used for example in the evaluation of the effect of physical exercise, or the response cold stress test.
Cyclic thermography is characterized by the repetition of the stimulation along the time. In this way, it is possible to achieve an accurate time-response, even if the thermal amplitudes are very small.
Preparing an Image Acquisition Setup
To perform a good thermal evaluation, it is extremely important to use not only the adequate modality but also to have a good preparation and appropriate setup. This should include a steady support for the camera, a clean table to manipulate the animal with high emissivity and low reflection and conductivity.
Firstly, a camera with the appropriate lenses should be employed. There are two types of thermal cameras (cooled and uncooled), thus the first step is to select the appropriate technology. Cooled cameras tend to be considerably more expensive but are more precise, and their sensitivity can go up to 13 mK. However, for most applications, a sensibility of 50 mK (provided by the vast majority of uncooled cameras) is sufficient. On the other hand, a very important aspect is the spatial resolution. Since the areas to observe in rats and mice are often very small, it is important to have an image with a high spatial resolution, to allow observation of small details after image acquisition. An image resolution of 320×240 pixels or higher is advisable. The lenses are also an important detail and will determine the distance to the focus area. A lens with a wide angle will be able to view a large area but not the small details and will lead to unfocussed images, as those presented in Figure 8. On the other hand, a lens with a small angle may require a far greater working space, by requiring a higher distance to view the same area. Small viewing areas also tend to create difficulties in interpreting the image and centering the object.
A relevant aspect in a thermal image is the thermal patterns, specially the natural body symmetry. Usually relative temperature (left compared to right) usually provide more information than the absolute temperature values. Another important aspect is to consider the average temperature in small areas, instead of evaluating a single point, which will necessarily conduct to wide errors.
Thermography in Animal Models of Cancer
Over the last 60 years, thermography has been used in clinical and pre-clinical (animal) studies (Ring, Hartmann, Ammer, Thomas, Land, Hand, 2010). Temperature changes at the surface of the skin may result from several biological phenomena, like: exercise (Aughey, Goodman, McKenna, 2014), articulation disorders, tumor development (Bendele, 2001; Magnan, Bondi, Pierantoni, Samaila, 2014) and drug delivery (Snekhalatha, Anburajan, Venkatraman, Menaka, 2013).
Heat and cold in a hot-blooded mammal can be either a diagnostic tool or a treatment procedure. Typically, a temperature increment or a left/right difference higher than 0.5 K is sufficient to require a more precise and complete diagnostic (Vardasca, 2012). However, for treatment purposes, a temperature variation of approximately 12 K is usually required (Portela et al., 2013, Rodrigues et al., 2013).
The thermographic evaluation of cancer patients usually takes place in hospitals, under thermoneutral temperature at approximately at 23 ºC. In order to develop reliable thermographic animal cancer models, such comfort conditions must be mimicked. However, for mice the thermoneutrality should be achieved at temperatures between 30 and 32 ºC (Gordon, 2012), which is far above the usual temperature in animal facilities. Other factors such the gender and the menstrual cycle of animals may also affect temperature readings (Renault et al., 2006).
The dynamics of angiogenesis in cancer have been intensively studied with recourse to animal models. Different models have allowed researchers to explore many features of cancer-associated angiogenesis, as well as to develop and test anti-angiogenic drugs. However, data from animal models must be carefully interpreted and extrapolation to human patients is not straightforward. For instance, clinical results obtained with anti-angiogenic drugs are often below what could be anticipated based on animal studies (Carmeliet & Jain, 2011). Choosing – and often, creating – adequate animal models to study specific aspects of angiogenesis is a critical step for planning translational research experiments.
Monitoring angiogenesis in animal models of cancer is also challenging. Post-mortem evaluations can be readily performed histologically, or by employing immunohistochemical or immunofluorescent techniques to detect and quantify blood or lymphatic vessels in tissue samples. In vivo monitoring of cancer-associated angiogenesis presents other difficulties. In this context, thermography offers its well-known advantages as a safe and non-invasive method. Several experiments have been performed using thermography to monitor physiological parameters and pathological changes in laboratory animals. However, only a few reports refer the use of thermography to monitor tumor growth and/or response to therapy in animal models of cancer.
From the point of view of physiology, thermographic techniques have been used to study the effects of dietary components on thermogenesis (Smriga, Murakami, Mori, Torii, 2000) and to correlate body temperature and cage behavior in rats (Bilodeau, 2011). Recently, David et al. (2013) used thermography to study the effects of environmental temperature in animal facilities on the metabolic status of mice. This study pointed thermography as an ideal tool for assessing husbandry practices and manage thermal stress in laboratory animals.
Thermography has also been used to study pathological phenomena, like the effects of sciatic nerve crush on the thermal status of the affected limb (Sacharuk et al., 2011). Thermography also allowed the detection of experimentally-induced pneumotorax in rats by demonstrating the presence of cooler thoracic areas Rich et al. (2004) and was shown to be useful in studying the viability of surgical skin flaps (Shejbal, Drvis, Bedekovic, 2012). In a rat model of serotonin-induced itch, thermography was used to study the local vasoregulation in response to serotonin administration, and a negative correlation between serotonin dose and local temperature was observed (Jasemian, Gazerani, Dagnaes-Hansen, 2012). In a recent study, Snekhalatha et al. (2013) showed that thermography may be used to monitor the development of arthritis in an experimental rat model and that thermographic findings significantly correlate with radiographic and other clinical data.
The need for adequate animal models to develop thermographic cancer research was recognized as early as 1986 (Jochimsen, Folk, Sundell, Loh, 1986). Two years later, Konerding and Steinberg (1988), reported the use of thermography to study subcutaneous (heterotopic) tumor xenografts in nude mice (Figure 10). Nude mice are particularly attractive for thermographic studies, since even a moderately thick hair coat will produce dramatic radiation scattering effects as well as change the skin temperature. Their extensive use in cancer biology for developing heterotopic (almost always subcutaneous) tumor xenografts is another reason why these models are so promising from the point of view of thermography.
|Figure 10. Nude mouse showing subcutaneous colon cancer xenograft (circle). The tumor temperature varies between 33.1 and 33.9 ºC (33.3 ± 0.1, average ± standard deviation) For the adjacent tissue, the average temperature is 34.1 ± 0.1|
An identical approach was used by Xie et al. (2004) to study the vascularization of breast cancer xenografts in mice. Both studies concluded that tumor xenografts showed reduced temperature, compared with the surrounding tissues. These findings were later confirmed by Song et al. (2007), who reported that tumor temperature progressively decreased as they developed, reaching a minimum of 3 ºC when compared with adjacent tissues, at 14 days post-implantation. These reports are in accordance with our own findings, depicted in Figure 10. The reason why mouse xenografts showed reduced temperature when compared with adjacent tissues, in stark contrast with previous thermographic in human cancer patients, is not fully understood. Mouse xenografts do not develop through the usual multistep carcinogenesis process, but are a population of, often, highly malignant cells, cultured in vitro and then injected into the animal. As such, it is possible that some tumors do not adequately represent the histological and, especially, vascular aspects of the original human tumors. Despite these surprising findings, early studies using mouse xenografts suggested that thermography could be a useful method to monitor tumor response to anticancer drugs in animal models (Xie, McCahon, Jakobsen, Parish, 2004; Song et al. 2007). Orthotopic rat xenografts (tumors or tumor cell lines transplanted into their corresponding organ of origin) are expected to be a more realistic model than heterotopic xenografts. One such model, mimicking colon carcinoma, is depicted in Figure 11.
|Figure 11. Nude rat with an orthotopic colon xenograft tumor. The tumor (AR01) shows an average temperature of 31.3 ºC, the subcutaneous tumor invasion area (AR02) 32.6 ºC and the adjacent tissue (AR03) 34.1 ºC|
However, thermography has seldom been adopted for this purpose. Recently, Rodrigues et al. (2013), reported on the successful use of thermography to monitor the response of mouse tumor xenografts treated with hyperthermia induced by magnetic nanoparticles. The same group also reported that thermographic data significantly correlated with readings obtained by intratumoral fiber-optic thermometers. A similar approach was successfully used by Tepper et al. (2013), for assessing the effects of diffusing alpha-emitters radiation therapy (DaRT) wires on subcutaneous breast cancer xenografts in mice. Intratumoral temperature differences were higher in treated versus untreated tumors, which agrees with the hypothesis that DaRT wires induced localized tumor destruction at implantation sites. The authors also found that tumor area estimated based on thermographic images was significantly correlated with direct visual caliper measurements. In a recent study, Hashida et al. (2014) successfully used a thermographic approach to monitor the thermal ablation of subcutaneous colon cancer xenografts using modified carbon nanotubes. Given the importance of nude mice xenografts in developing anti-cancer therapies, and the high costs associated with other imaging technologies, it is expected that thermography, with its inherent advantages, will attract the attention of researchers working with this kind of models in the near future.
When using hirsute animals, clipping the hair that overlies the tumor area is necessary in order to obtain suitable readings (Figure 12). This problem is frequently encountered when employing common rat strains (e.g. Fischer or Wistar rats). Poljak-Blazi et al. (2009) studied thermographically the effects of experimental tumor transplantation, local inflammation and hematoma in Sprague-Dawley rats. As expected, the authors observed a raise in skin temperature following inflammation or hematoma induction. As observed in mouse xenografts, transplanted tumors (Walker 256 carcinoma) showed reduced temperature compared with adjacent tissues. However, contrary to what has been described for mice, this trend changed from day 10 post-transplantation onwards, when tumor temperature increased up to 1 ºC over adjacent tissues. The authors reported that, histologically, the tumors showed abundant vascularization, which may, to some degree, explain these findings. A more recent study addressed the relationship between thermographic findings and tumor vascularization, as detected by ultrasonography, in nitrosamine-induced mammary cancer (Faustino-Rocha, 2013). In this study, significant intratumoral thermal heterogeneity was observed. Maximum intratumoral temperature and thermal amplitude were found to correlate positively with tumor size. This may reflect tumor heterogeneity, typically found in large lesions, which partially agrees with the findings of Poljak-Blazi et al. (2009), suggesting that, at least in rats, long-term tumor progression leads to increasing tumor temperatures. However, the comparison between the results of Faustino-Rocha et al. (2013) and of Poljak-Blazi et al. (2009) is necessarily limited, because the previous study does not compare tumor temperature with that of adjacent tissues. Concerning the relationship between tumor temperature and vascularization, the authors reported that maximum tumor temperature and thermal amplitude were also correlated with blood vessel density, as determined using an ultrasonographic technique (power Doppler). A more recent study came to confirm these findings, Faustino-Rocha et al. (2016) show that thermal readings may be correlated with tumor vascularization, as assessed histologically and immunohistochemically (VEGF immunoexpression).
|Figure 12. N-methyl-N-nitrosourea-induced model of mammary cancer in a female Wistar rat. Visible (a) and infrared thermal images using two different temperature scales (b and c) provide complementary information regarding the tumor (circled area)|
The thermal differences that have been observed between tumors in different animal models certainly reflect important differences in tumor development and biology. Over recent years, it has become evident that mouse xenografts only mimic the original human tumor to a limited extent (Eklund, Bry, Alitalo, 2013). Importantly, it has been found that anti-angiogenic agents which are highly effective in mouse xenografts show comparatively low efficacy in human patients, who present a more complex scenario. As a result, there is a search for models that recapitulate multistep carcinogenesis more accurately, including the progressive development of tumor vasculature (Eklund, Bry, Alitalo, 2013). A detailed and systematic characterization of such models is also mandatory. In this setting, thermography, with its inherent advantages over other invasive, expensive, or radiation-emitting techniques, maintains all its untapped potential to become a standard technique for monitoring tumor angiogenesis in animal models.
Thermography is a viable technique to assess the vascularization of superficial (cutaneous or subcutaneous) tumors in laboratory animals. As demonstrated by various studies, it may complement other techniques, such as ultrasound imaging and histopathology, providing relevant information on tumor biology. However, the presence of a dense haircoat on many mouse and rat strains, as well as the presence of a thick subcutaneous fat layer (e.g. in pigs), may limit the use of thermography. Other limitation of this technique when used in laboratory rodents is the frequent need for animal handling, restrainment or sedation, which may influence thermal readings. Also, it is not always possible to obtain thermal images from a 90º angle, as recommended, due to the animal’s shape.
Finally, thermal cameras suppliers have recently started offering devices, which make it possible to obtain simultaneous and superimposable thermal and visible images, which will add to the potential of this technology. At the same time, the continuous development of smaller thermal cameras also offers new possibilities and applications.
This research was previously published in Innovative Research in Thermal Imaging for Biology and Medicine edited by Ricardo Vardasca and Joaquim Gabriel Mendes, pages 237-263, copyright year 2017 by Medical Information Science Reference (an imprint of IGI Global).
The authors would like to acknowledge to Project POCI-01-0145-FEDER-006939, LEPABE and POCI-01-0145-FEDER-006958, funded by FEDER funds through COMPETE2020 – POCI – and by Portuguese Government and the Social European Fund through FCT, grant SFRH/BPD/85462/2012, UID/EMS/50022/2013(LAETA), UID/AGR/04033/2013, PTDC/DES/114122/2009, SFRH/BD/102099/2014, PEst-C/SAU/UI3282/2013 and IBILI-FMUC through PEst-C/SAU/UI3282/2013, UID/NEU/04539/2013 and ACIMAGO.
Aughey, R. J., Goodman, C. A., & McKenna, M. J. (2014). Greater chance of high core temperatures with modified pacing strategy during team sport in the heat. Journal of Science and Medicine in Sport , 17(1), 113–118. doi:10.1016/j.jsams.2013.02.013
Bartling, S. H., Stiller, W., Semmler, W., & Kiessling, F. (2007). Small animal computed tomography imaging. Current Medical Imaging Reviews , 3(1), 45–59. doi:10.2174/157340507779940327
Bendele, A. (2001). Animal models of rheumatoid arthritis. Journal of Musculoskeletal & Neuronal Interactions , 1(4), 377–385.
Bignold, L. P. (2003). Initiation of genetic instability and tumour formation: A review and hypothesis of a nongenotoxic mechanism. Cellular and Molecular Life Sciences , 60(6), 1107–1117. doi:10.1007/s00018-003-2314-6
Bignold, L. P., Coghlan, B. L. D., & Jersmann, H. P. A. (2006). Cancer morphology, carcinogenesis and genetic instability: a background. In Cancer: Cell Structures, Carcinogens and Genomic Instability (pp. 1-24). Birkhäuser Basel. doi:10.1007/3-7643-7378-4_1
Bilodeau, G. A., Ghali, R., Desgent, S., Langlois, P., Farah, R., St-Onge, P. L., . . . Carmant, L. (2011, June). Where is the rat? Tracking in low contrast thermographic images. In CVPR 2011 Workshops (pp. 55-60). IEEE. doi:10.1109/CVPRW.2011.5981685
Carmeliet, P., & Jain, R. K. (2011). Molecular mechanisms and clinical applications of angiogenesis. Nature , 473(7347), 298–307. doi:10.1038/nature10144
Chatham, J. C., & Blackband, S. J. (2001). Nuclear magnetic resonance spectroscopy and imaging in animal research. ILAR Journal , 42(3), 189–208. doi:10.1093/ilar.42.3.189
Cross, S. (2013). Underwood’s Pathology (6th ed.). Elsevier Health Sciences.
David, J. M., Chatziioannou, A. F., Taschereau, R., Wang, H., & Stout, D. B. (2013). The hidden cost of housing practices: Using noninvasive imaging to quantify the metabolic demands of chronic cold stress of laboratory mice. Comparative Medicine , 63(5), 386–391.
Decker, S., & Sausville, E. (2011). Perspectives on the History and Evolution of Tumor Models . In Tumor Models in Cancer Research (pp. 3–20). Humana Press. doi:10.1007/978-1-60761-968-0_1
Dranoff, G. (2012). Experimental mouse tumour models: What can be learnt about human cancer immunology? Nature Reviews. Immunology , 12(1), 61–66.
Eklund, L., Bry, M., & Alitalo, K. (2013). Mouse models for studying angiogenesis and lymphangiogenesis in cancer. Molecular Oncology , 7(2), 259–282. doi:10.1016/j.molonc.2013.02.007
Farago, A. F., Snyder, E. L., & Jacks, T. (2012). SnapShot: Lung cancer models. Cell , 149(1), 246–246. doi:10.1016/j.cell.2012.03.015
Faustino-Rocha, A. I., Silva, A., Gabriel, J., da Costa, R. G., Moutinho, M., Oliveira, P. A., & Ginja, M. (2016). Long-term exercise training as a modulator of mammary cancer vascularization. Biomedicine and Pharmacotherapy , 81, 273–280. doi:10.1016/j.biopha.2016.04.030
Faustino-Rocha, A. I., Silva, A., Gabriel, J., Teixeira-Guedes, C. I., Lopes, C., da Costa, R. G., & Ginja, M. (2013). Ultrasonographic, thermographic and histologic evaluation of MNU-induced mammary tumors in female Sprague-Dawley rats. Biomedicine and Pharmacotherapy , 67(8), 771–776. doi:10.1016/j.biopha.2013.06.011
Frese, K. K., & Tuveson, D. A. (2007). Maximizing mouse cancer models. Nature Reviews. Cancer , 7(9), 654–658. doi:10.1038/nrc2192
Gaivão, F. D. M., de Lima, J. P., Agostinho, A. G., Oliveira, A. S., & Peres, A. P. (2001). Imagiologia Clínica-Princípios e Técnicas. Academic Press.
Gordon, C. J. (2012). Thermal physiology of laboratory mice: Defining thermoneutrality. Journal of Thermal Biology , 37(8), 654–685. doi:10.1016/j.jtherbio.2012.08.004
Grossfeld, G. D., Ginsberg, D. A., Stein, J. P., Bochner, B. H., Esrig, D., Nichols, P. W., & Cote, R. J. (1997). Thrombospondin-1 expression in bladder cancer: Association with p53 alterations, tumor angiogenesis, and tumor progression. Journal of the National Cancer Institute , 89(3), 219–227. doi:10.1093/jnci/89.3.219
Hanahan, D., Wagner, E. F., & Palmiter, R. D. (2007). The origins of oncomice: A history of the first transgenic mice genetically engineered to develop cancer. Genes & Development , 21(18), 2258–2270. doi:10.1101/gad.1583307
Hansen, K., & Khanna, C. (2004). Spontaneous and genetically engineered animal models: Use in preclinical cancer drug development. European Journal of Cancer , 40(6), 858–880. doi:10.1016/j.ejca.2003.11.031
Harshaw, C., Culligan, J. J., & Alberts, J. R. (2014). Sex differences in thermogenesis structure behavior and contact within huddles of infant mice. PLoS ONE , 9(1), e87405. doi:10.1371/journal.pone.0087405
Hashida, Y., Tanaka, H., Zhou, S., Kawakami, S., Yamashita, F., Murakami, T., & Hashida, M. (2014). Photothermal ablation of tumor cells using a single-walled carbon nanotube–peptide composite. Journal of Controlled Release , 173, 59–66. doi:10.1016/j.jconrel.2013.10.039
Iacobuzio-Donahue, C. A. (2009). Epigenetic changes in cancer. Annual Review of Pathological Mechanical Disease , 4(1), 229–249. doi:10.1146/annurev.pathol.3.121806.151442
Jasemian, Y., Gazerani, P., & Dagnaes-Hansen, F. (2012). Infrared thermography in serotonin-induced itch model in rats. Open Dermatology Journal , 6(1), 1–7. doi:10.2174/1874372201206010001
Jochimsen, P. R., Folk, G. E., Sundell, M., & Loh, P. M. (1986). B16 melanoma in hairless mice: A model for thermographic research. Journal of Surgical Oncology , 33(1), 50–52. doi:10.1002/jso.2930330114
Jonkers, J., & Berns, A. (2005). Animal models of cancer: Introduction to the cellular and molecular biology of cancer . Oxford University Press.
Konerding, M. A., & Steinberg, F. (1988). Computerized infrared thermography and ultrastructural studies of xenotransplanted human tumors on nude mice. Thermology , 3(1), 7–14.
Lima, J. J. D. (2011). Nuclear Medicine Physics. Series in Medical Physics and Biomedical Engineering . Taylor and Francis.
Lima, J. J. P. D. (1995). Física dos métodos de imagem com raios X . Coimbra: Edições ASA.
Loi, M., Di Paolo, D., Becherini, P., Zorzoli, A., Perri, P., Carosio, R., & Ponzoni, M. (2011). The use of the orthotopic model to validate antivascular therapies for cancer. The International Journal of Developmental Biology , 55(4-5), 547–555. doi:10.1387/ijdb.103230ml
MacEwen, E. G. (1990). Spontaneous tumors in dogs and cats: Models for the study of cancer biology and treatment. Cancer and Metastasis Reviews , 9(2), 125–136. doi:10.1007/BF00046339
Magnan, B., Bondi, M., Pierantoni, S., & Samaila, E. (2014). The pathogenesis of Achilles tendinopathy: A systematic review. Foot and Ankle Surgery , 20(3), 154–159. doi:10.1016/j.fas.2014.02.010
Merck Sharp & Dohme Corp. (2014). The Merck Veterinary Manual – Diagnostic Imaging, Radiology. Retrieved from www.merckmanuals.com/vet/clinical_pathology_and_procedures/diagnostic_imaging/radiography.html
Oliveira, P. A., Colaço, A., Chaves, R., Guedes-Pinto, H., De-La-Cruz, P., Luis, F., & Lopes, C. (2007). Chemical carcinogenesis. Anais da Academia Brasileira de Ciencias , 79(4), 593–616. doi:10.1590/S0001-37652007000400004
Oliveira, P. A., Colaço, A., De la Cruz, P. L., & Lopes, C. (2006). Experimental bladder carcinogenesis-rodent models. Experimental Oncology , 28(1), 2–11.
Polin, L., Corbett, T. H., Roberts, B. J., Lawson, A. J., Leopold, W. R., III, White, K., . . . Horwitz, J. P. (2011). Transplantable Syngeneic Rodent Tumors: Solid Tumors in Mice. In Tumor Models in Cancer Research (pp. 43-78). Humana Press.
Poljak-Blazi, M., Kolaric, D., Jaganjac, M., Zarkovic, K., Skala, K., & Zarkovic, N. (2009). Specific thermographic changes during Walker 256 carcinoma development: Differential infrared imaging of tumour, inflammation and haematoma. Cancer Detection and Prevention , 32(5), 431–436. doi:10.1016/j.cdp.2009.01.002
Portela, A., Vasconcelos, M., Fernandes, M. H., Garcia, M., Silva, A., Gabriel, J., & Cavalheiro, J. (2013). Highly focalised thermotherapy using a ferrimagnetic cement in the treatment of a melanoma mouse model by low temperature hyperthermia. International Journal of Hyperthermia , 29(2), 121–132. doi:10.3109/02656736.2013.767478
Potente, M., Gerhardt, H., & Carmeliet, P. (2011). Basic and therapeutic aspects of angiogenesis. Cell , 146(6), 873–887. doi:10.1016/j.cell.2011.08.039
Renault, G., Bonnin, P., Marchiol-Fournigault, C., Gregoire, J. M., Serriere, S., Richard, B., & Fradelizi, D. (2006). High-resolution ultrasound imaging of the mouse. Journal de Radiologie , 87(12 Pt 2), 1937–1945. doi:10.1016/S0221-0363(06)74179-8
Rich, P. B., Dulabon, G. R., Douillet, C. D., Listwa, T. M., Robinson, W. P., Zarzaur, B. L., & Katz, L. M. (2004). Infrared thermography: A rapid, portable, and accurate technique to detect experimental pneumothorax. The Journal of Surgical Research , 120(2), 163–170. doi:10.1016/j.jss.2004.02.026
Ring, E. F. J., Hartmann, J., Ammer, K., Thomas, R., Land, D., & Hand, J. W. (2010). Infrared and Microwave Medical Thermometry. Experimental Methods in the Physical Sciences , 43, 393–448. doi:10.1016/S1079-4042(09)04307-0
Rodrigues, H. F., Mello, F. M., Branquinho, L. C., Zufelato, N., Silveira-Lacerda, E. P., & Bakuzis, A. F. (2013). Real-time infrared thermography detection of magnetic nanoparticle hyperthermia in a murine model under a non-uniform field configuration. International Journal of Hyperthermia , 29(8), 752–767. doi:10.3109/02656736.2013.839056
Rowland, D. J., & Cherry, S. R. (2008, May). Small-animal preclinical nuclear medicine instrumentation and methodology. Seminars in Nuclear Medicine , 38(3), 209–222. doi:10.1053/j.semnuclmed.2008.01.004
Ryer, A., & Light, V. (1997). Light measurement handbook. Academic Press.
Sacharuk, V. Z., Lovatel, G. A., Ilha, J., Marcuzzo, S., Pinho, A. S. D., Xavier, L. L., & Achaval, M. (2011). Thermographic evaluation of hind paw skin temperature and functional recovery of locomotion after sciatic nerve crush in rats. Clinics (Sao Paulo) , 66(7), 1259–1266. doi:10.1590/S1807-59322011000700023
Shejbal, D., Drvis, P., & Bedekovic, V. (2012). Thermography‐measured effect of capsaicin, methylprednisolone and mitomycin on the survival of random skin flaps in rats. Skin Research and Technology , 18(2), 157–161. doi:10.1111/j.1600-0846.2011.00547.x
Shojaei, F. (2012). Anti-angiogenesis therapy in cancer: Current challenges and future perspectives. Cancer Letters , 320(2), 130–137. doi:10.1016/j.canlet.2012.03.008
Singh, M., Murriel, C. L., & Johnson, L. (2012). Genetically engineered mouse models: Closing the gap between preclinical data and trial outcomes. Cancer Research , 72(11), 2695–2700. doi:10.1158/0008-5472.CAN-11-2786
Smriga, M., Murakami, H., Mori, M., & Torii, K. (2000). Use of thermal photography to explore the age-dependent effect of monosodium glutamate, NaCl and glucose on brown adipose tissue thermogenesis. Physiology & Behavior , 71(3), 403–407. doi:10.1016/S0031-9384(00)00350-4
Snekhalatha, U., Anburajan, M., Venkatraman, B., & Menaka, M. (2013). Evaluation of complete Freunds adjuvant-induced arthritis in a Wistar rat model. Zeitschrift fur Rheumatologie , 72(4), 375–382. doi:10.1007/s00393-012-1083-8
Song, C., Appleyard, V., Murray, K., Frank, T., Sibbett, W., Cuschieri, A., & Thompson, A. (2007). Thermographic assessment of tumor growth in mouse xenografts. International Journal of Cancer , 121(5), 1055–1058. doi:10.1002/ijc.22808
Tepper, M., Shoval, A., Hoffer, O., Confino, H., Schmidt, M., Kelson, I., & Gannot, I. (2013). Thermographic investigation of tumor size, and its correlation to tumor relative temperature, in mice with transplantable solid breast carcinoma. Journal of Biomedical Optics , 18(11), 111410–111410. doi:10.1117/1.JBO.18.11.111410
Torre, L. A., Bray, F., Siegel, R. L., Ferlay, J., Lortet‐Tieulent, J., & Jemal, A. (2015). Global cancer statistics, 2012. CA: a Cancer Journal for Clinicians , 65(2), 87–108. doi:10.3322/caac.21262
Van Middendorp, J. J., Sanchez, G. M., & Burridge, A. L. (2010). The Edwin Smith papyrus: A clinical reappraisal of the oldest known document on spinal injuries. European Spine Journal , 19(11), 1815–1823. doi:10.1007/s00586-010-1523-6
Vardasca, R. Â. R. (2012). The effect of work related mechanical stress on the peripheral temperature of the hand. Academic Press.
Xie, W., McCahon, P., Jakobsen, K., & Parish, C. (2004). Evaluation of the ability of digital infrared imaging to detect vascular changes in experimental animal tumours. International Journal of Cancer , 108(5), 790–794. doi:10.1002/ijc.11618