Josh P. Roberts has an M.A. in the history and philosophy of science, and he also went through the Ph.D. program in molecular, cellular, developmental biology, and genetics at the University of Minnesota, with dissertation research in ocular immunology.
There are many reasons for running molecular imaging studies of small animals, but perhaps the most significant is the ability to conduct longitudinal studies, which follow the progression of disease in individual animals.
“The old-fashioned way of ‘kill and grind’ and figure out what’s going on at a given time point doesn’t really tell you the big story,” says Timothy Doyle, scientific director of the Stanford Small Animal Imaging Facility. With longitudinal imaging studies, each animal effectively serves as its own control. Experiments can be refined on the go, if necessary. And the number of animal subjects, along with their associated costs, is greatly reduced.
Broadly speaking, pre-clinical imaging modalities (like their clinical counterparts) fall into two categories. Anatomical or morphological imaging—exemplified by x-ray—documents the physical structure. Molecular imaging—exemplified by nuclear techniques such positron emission tomography (PET) scans—uses probes to detect particular biological targets or pathways.
Researchers today often combine anatomical with molecular techniques, reconstruct 3D images from series of scans and use related modalities to determine the location of cancer metastases and the size of tumors following treatment.
Here we look at current tools for small-animal whole-body imaging and take a peek into the technology’s future.
Anatomical imaging
Anatomical imaging is the oldest imaging modality, says Alexandra De Lille, director of technical applications, in vivo imaging, at Revvity. “We’re mostly talking about x-ray, which became more sophisticated with CT [computed-tomography] imaging, the 3D reconstruction of x-rays.”
Just as your dentist probably ditched x-ray film for lower-power, higher-resolution digital alternatives, today’s small-animal x-ray imagers are smaller and faster than ever, allowing mice to be whole-body scanned in tens of seconds at about 100-micron resolution, with relatively low x-ray doses, Doyle notes.
CT excels at imaging hard tissue like bone. But soft tissues, such as organs, the brain, and small tumors, are more difficult to visualize. “CT has historically been the anatomical modality to be used in multimodal imaging. But it gives only bone structure, not soft tissue. Especially not good for brain, whose skull absorbs most of the x-rays,” says van Cauter.
Researchers can use magnetic resonance imaging (MRI) to visualize any organ or tissue in situ, says Staf van Cauter, vice president of sales, marketing and business development at TriFoil Imaging.
MRIs typically are large, expensive pieces of liquid helium-cryocooled equipment that require extensive shielding of the electromagnetic field via a Faraday cage. But smaller instruments also are available for the small-animal imaging market.
Players like Siemens and Agilent have recently exited the pre-clinical market, but several others have come out with diminutive MRIs for small animals, including Bruker Biospin. These have permanent magnets that require no cryocooling and are self-shielded, meaning that they can be placed anywhere a CT is housed, including on a benchtop.
Ultrasound also is used for anatomical imaging. It and MRI—especially with the use of contrast agents—are the modalities of choice for real-time functional studies of, for example, blood flow and cardiac function.
Molecular imaging
Molecular imaging modalities typically use either a radioactive tracer or light emission to indicate the presence of a molecule or process of interest.
In nuclear modalities such as PET and its relative, single photon emission computed tomography (SPECT), gamma rays emitted by tracer molecules such as 18F-fluorodeoxyglucose and 11C-choline are detected by cameras positioned at different angles, or rotated around the animal, to create a three-dimensional picture of where in the body the tracer is localized.
Because tissue is effectively transparent to gamma radiation, nuclear modalities are ideal for even deep tissue imaging. Yet the very fact that radioactive materials are involved has significant implications in terms of animal husbandry and radiation control, says De Lille. That, coupled with the need to coordinate the experimental time line with the availability of the probe, means that “organizationally it becomes extremely challenging from the moment you have to work with rapidly decaying radionuclides,” she adds.
Researchers also can use optical modalities, which typically capture light either from genetically encoded reporters (such as luciferase) or from fluorescently labeled probes. These modalities typically are limited to depths of about a centimeter, which can be a limitation in some models but “works great for mice,” says Doyle. “If it’s too deep from one view, you just turn the animal over and view from the other side, or do some sort of 3D imaging.”
Revvity supports bioluminescence, which Doyle calls “probably the biggest modality for pre-clinical imaging right now,” with its popular IVIS platform. Bioluminescence is incredibly sensitive, enabling researchers to see metastases as small as tens of cells “long before you could determine it with any other sort of modality,” Doyle says. But the technique requires luciferase expression, which is typically performed by making a transgenic animal or adoptive transfer of luciferase-expressing cells, which doesn’t translate easily to human patients.
Fluorescence is more translatable, and also a popular application for small-animal imaging, Doyle notes. Tissues’ strong absorbance of visible light, combined with inherent autofluorescence, traditionally has hampered the use of fluorescence for in vivo imaging. Yet in the near-infrared (NIR) range—between 650 nm and 900 nm—“the animal becomes semi-transparent,” says van Cauter, and there is a trend in the industry to create fluorescent probes in this region. Indeed, one fluorophore—Indocyanine Green (ICG)—already has been approved for clinical use (though it lacks a conjugation group).
Two- and three-dimensional fluorescence instruments are available from several vendors, Doyle says. Applications “haven’t really taken off yet, but they’re starting to get some excitement.”
Make it a combo
That said, researchers rarely rely on a single modality. They often combine them, either in a single instrument or a series of compatible instruments, from which fiducial marks can be registered and data overlaid to generate 3D multimodal images. For example, TriFoil—a new vendor comprising pre-clinical assets from GE, Gamma Medica and Bioscan—offers instruments that pair PET, SPECT and fluorescence emission computed tomography (FLECT) with CT or MRI in various combinations. A CT scanner can be combined in the same box with the IVIS system, with the same animal holder (shuttle) compatible with Sofie PET scanners (which Revvity is now distributing). And Bruker offers nine different pre-clinical modalities—including optical, MRI and magnetic particle imaging (MPI)—which share either a common instrument or shuttle.
Researchers need to consider the relative importance of such factors as capital, operating and regulatory costs, footprint, resolution and sensitivity, translatability to the clinic, signal depth and the need for 3D localization when choosing a pre-clinical imaging modality. Fortunately, where one modality may fall short another is there—sometimes even in the same box—to pick up the slack.