There is increasing consensus that “the majority of medical interventions introduced into clinical development prove unsafe or ineffective.” Clearly, the research community needs to do a better job with studies that are meant to guide the development of novel treatments for human diseases.
Standardizing the quality of published preclinical data should weed out irreproducible results that are costly and have weak chances of translational success. Too often clinical studies based on flawed preclinical data bring false hope to the families involved.
We can improve the success rate for finding useful treatments to various human diseases by ensuring that our studies meet the highest standards of scientific rigor and transparency.
What needs to change?
Too many scientific publications lack key information on how the study was designed, conducted and analyzed. Way too often the data from animal experiments cannot be fully evaluated, replicated or utilized. This clearly limits their value in informing future scientific studies, designing drugs and making policy decisions.
For scientific, ethical and economic reasons, experiments involving animals need to be correctly designed, appropriately analyzed and transparently reported. This increases the scientific validity of the results and maximizes the knowledge gained from each experiment.
Below are some fundamental steps to develop robust experimental designs when using mouse models of human diseases.
Studies should be planned before they are started. The experimental design changes depending on the objectives of the study. A well designed experiment should be powerful enough to detect biologically relevant effects.
1. Minimize bias
Be cognizant of the potential for bias when assessing results. It is critically important to minimize bias when assessing results, especially if there is any subjective element in assessing the outcomes.
To avoid bias, experiments should be performed so that researchers are “blind” to the allocation of animals to treatment groups and mouse genotypes.
Be sure to report the blinding conditions when publishing data to increase study transparency and reproducibility.
Table 1. Additional guidelines for the design and publication of preclinical experiments using mouse models of human diseases.
2. Housing and husbandry considerations
If your study involves mice with different genotypes, attempt to generate all genotypes in the same litter to ensure comparable environment and maternal influences on your experimental mice.
Distribute mice from each litter across different experimental groups. Good practice involves weaning animals at the same age into cages containing mice from multiple litters segregated by genotype.
Because a number of phenotypes depend on the health and the immune status of the mice, it is crucial that researchers report these variables together with the environmental enrichment conditions to facilitate the reproducibility of efficacy studies.
3. Report methods and results accurately
Report baseline data prior to treatment or testing (characterize inclusion/exclusion criteria, disease severity, age, etc.). It is important that you report dose-response results. Also, verify that the interventional drug or biologic reaches and engages the target.
Keep in mind that preclinical research should be replicated in multiple models of the same disease and in different species, and also should be replicated independently, before moving on to clinical trials.
Report your results clearly and simply—save everyone time and effort (and the pain of having to read between the lines). Spell out the relevant information to ensure that the methods and results of your study can be reviewed, analyzed and repeated by other investigators.
Check out the 20-point checklist of the essential information that should be included in publications reporting animal research crafted by Animal Research Reporting of In Vivo Experiments (ARRIVE). These guidelines are aimed at scientists who are seeking funding, writing their research for publication, or taking part in the peer-review process.
Funding agencies also are demanding significant improvements in the quality of the preclinical and clinical research they support. For a great example see: NINDS expects rigorous study design and transparent reporting.
Too many of these recommendations are shockingly not implemented in preclinical research. It is unclear whether the failure to implement these guidelines explains the recurrent discordance of results between drug safety and efficacy obtained in preclinical stages and clinical trials.
With the recent outpouring of numerous field guides for working with mice that model human diseases, there is great optimism that the preclinical culture is turning the corner to produce more clinically relevant results.
Consistent adherence to robust preclinical experimental designs will improve the process from bench to bedside in the interest of patients who greatly need effective treatments.
As an example, many different mouse models have been created for Huntington’s disease research and development. Based on your research focus, the choice of the model(s) to use is important and should be considered thoroughly.