Preclinical Animal Models in Cancer Research and Drug Development
Preclinical animal models are used in cancer research for two principal reasons. One is to enable us to examine biological models which are physiologically representative of the human condition, and to use these to improve our understanding of the clinical situation.
Models are also used as part of a therapeutic development portfolio. They allow us to establish whether a novel agent reaches its site of action at an achievable concentration for significant antitumor efficacy, without showing unreasonable toxicity.
Conventional Model Types
Subcutaneous cell line models help us assess the therapeutic potential of novel agents. However, with tumors growing as a defined mass on the flank of an animal, they often don’t physiologically represent disease.
PDX models, where tissues are derived from patient samples rather than cell lines, are an improvement and do better recapitulate the clinical morphology of cancer. It is, however, rare to find metastatic involvement when using subcutaneous models .
Orthotopic xenograft models, where implanted cells are grown in their respective clinically relevant host tissues, allow for the development of a more representative tumor microenvironment. They also potentially offer a more faithful pattern of disease spread and metastasis.
Systemic Tumor Models
When we look at systemic (or “liquid”) cancers, such as leukemia, growing these as solid tumors has little clinical relevance. Classically, leukemias have been implanted directly into the circulatory blood via tail vein or direct cardiac injection. However, this type of implantation is inherently inconsistent with varied growth and take rates.
More recently, techniques have been developed to allow direct implantation of cells into the bone marrow compartment via direct intratibial or intrafemoral injection. This places leukemic cells in their natural environment and leads to much more consistent engraftment and growth.
Improvements in Recipient Animal Strains
Nude or athymic mice lack functional T cells, but residual immunity in these strains limits their ability to engraft human leukemic cells. Human leukemic engraftment couldn’t really be meaningfully studied in preclinical models until the development of more immunocompromised mouse strains. This includes models such as the non-obese diabetic, severely-compromised immunodeficient (NOD/SCID) mouse which lacks T and B cells .
Improved NOD/SCID mouse strains such as the NSG™ and the NOG® mouse have now been developed which lack both natural killer (NK) cell and macrophage activity . The further depletion of immunity in these animals allows more efficient engraftment of human leukemic cells than in conventional NOD/SCID mice .
Measuring Disease Burden of Non-Solid Cancers
Even allowing for improvements in implantation, measurement of disease burden in systemic models remains problematic, relying on terminal endpoints such as hind limb paralysis.
FACS analysis can be used with blood sampling or bone marrow puncture. Often however, disease does not show at measurable levels in peripheral blood until late stage. This means that if a researcher waits until disease is at a measurable level before treatment, there is only a very small therapeutic window.
When using bone marrow puncture, disease levels may vary throughout the bone marrow compartment. Therefore, results from bone marrow sampling can be inconsistent and can easily over- or underestimate the level of disease in an animal.
In both of these cases, it should also be noted that it is possible to alter the course of the disease while removing leukemic cells from the animal. This makes it hard to interpret the therapeutic effect of test items in the model.
As disease burden is difficult to measure, animal selection and randomization is challenging. To accommodate for this, larger groups of animals are needed to control for variability throughout the study.
Optical Imaging Technology
One way to address the challenges associated with systemic disease is to utilize optical imaging technology (which we’ve previously reviewed for overall preclinical applications). By implanting luciferase-labeled cells (luc+) orthotopically, we can model disease in a reliable and reproducible manner.
Bioluminescence is a biochemical reaction, where light is emitted when an enzyme is exposed to its substrate. Firefly luciferase, plus its substrate, luciferin, are most often used in optical imaging, though many other luciferases are also available. Usually, the subject is injected with luciferin immediately before imaging begins. When the luciferin reaches luciferase-expressing cells, light is emitted. This is detected and the signal correlated to disease burden.
Using state-of-the-art imaging techniques, we’re able to follow systemic disease throughout study duration, and provide insight into test item effects at an early stage of treatment.
Optical imaging also allows us to track both disease development and spread in individual animals throughout the whole course of a study. The presence and location of disease in animals can be confirmed immediately post implantation, before treatment starts, and at multiple time points over the course of disease development. This means that each animal essentially becomes its own control.
This allows for real time assessment of disease burden, rational randomization of study groups, and smaller group sizes. Imaging data can even be used to determine appropriate time points for FACS sampling if required.
In addition to bioluminescent imaging methods, a number of fluorophores have been developed which emit in the near infra-red portion of the spectrum (emission wavelengths >680nm). At these wavelengths in vivo tissue is basically transparent, making these fluorophores ideal for optical imaging at depth.
Fluorophores can be used to label a wide range of agents, such as antibodies and small molecule drugs. This gives us the ability to visualize a range of biological events taking place within a live animal in real time.
Optical Imaging of Leukemia Models
Optical imaging has been used in the development of several elegant leukemia models, allowing the study of multiple aspects of disease. This includes leukemic stem cells and competitive implantation, the spread of systemic disease from a single site of implantation, and therapeutic models to study the effect of novel agents [5,6].
Combining this data with other complementary imaging technologies (e.g. X-Ray, CT, or MRI) allows coregistration of anatomical and biological data enabling a deeper understanding of the mechanisms at play in systemic disease.
Optical imaging allows us to answer a range of biological questions about systemic disease in a rapid, sensitive, reliable, and cost-effective manner:
- Multiple biological questions can be answered in the same experiment by using a variety of bioluminescent and fluorescent markers.
- Systemic disease can be followed in real time, in live animals, with the ability to assess disease burden throughout the course of the study.
- The number of mice needed per study is reduced, with longitudinal follow up of each animal at several time points.
- The in vivo biodistribution of therapeutics such as antibodies, ADCs, nanoparticles, and small molecules can be visualized.
- Anatomical reference points can be provided alongside biological data through combination with other imaging modalities.
- Through combination with more conventional techniques such as FACS analysis, a fully comprehensive data set can be provided.
 Agliano et al. Human acute leukemia cells injected in NOD/LtSz/IL-2R gamma null mice generate a faster and more efficient disease compared to other NOD/SCID related strains. International Journal of Cancer 2008;123:2222–7.