9 Ways to Optimize Preclinical Imaging

Preclinical models and the modalities used within them are tools to study certain aspects of disease, such as neurodegeneration in diseases of the central nervous system. Following are tips to ensure imaging in a preclinical space generates more translatable data across all disease states, helping drug developers better design and monitor clinical trials and bring drugs to market faster.

1. Whenever possible, and prior to engaging in a study, connect preclinical imaging specialists with those from other areas within the development process (preclinical, translational, and clinical development) to evaluate the translatability of model components. The pertinence of the particular disease model and the aspects of the disease that are being examined should be taken into account. It may be of no use to examine aspects of the disease that would not have a translatable imaging aspect. Scaling from rodents to humans may not always be straightforward but the possible translational aspects of certain parameters—e.g., the metabolites from MRS (magnetic resonance spectroscopy)—should be explained.
2. Consider all aspects of a modeling strategy. Imaging is very often just one aspect of a modeling paradigm along with function, behavior, electrophysiological, and histopathological components. The paradigm should be examined with respect to the timing and results of these assays relative to imaging endpoints.
3. Ask for data. It is important to ensure, when and where possible, that prior data is available from the imaging group that can show that there is either a discernible deficit within the modeling paradigm or that the parameters being studied can actually be examined by the group performing the study.
4. Co-validate new paradigms to establish new methodologies and design studies in a cost-effective way. Again, data from the specific scanner being used and representative data and analysis from the team employing the scanner is important. Doing the same study in a different scanner can have an adverse effect on how results are interpreted.
5. Employ “test-retest” measures in order to estimate within group and within subject variations. The particular animal model and the specific imaging methodologies being applied and the data analysis selected may all affect the outcome. This is extremely important when possibly subtle changes upon drug treatment are of interest.
6. Discuss the possibilities as well as the limitations of the disease model in question. In the same context, discuss openly and decide beforehand the exclusion criteria (if any) and when/where/if to replace subjects in the study.
7. High throughput and high quality? Most definitely these terms can co-exist, but it is important to optimize the imaging time used per subject to extract only the relevant information with subsequent capability of retaining high throughput. Again, discussing methodologies with the experts can maintain expectations and allow for easier and quicker interpretation of data from imaging studies.
8. Use multiple time-point imaging to allow for better insight into disease pathology and drug effect or lack of effect.
9. Address all questions using a combination of methods—preferably from one scanning period within each study subject.