Module 9: Animal models and experimental design

Question 1

Describe the concepts of fidelity and discrimination, and discuss validation criteria for an induced and a spontaneous animal model.

Answer 1

• Fidelity= refers to how closely an animal model replicates the human disease
  –> High-fidelity models exhibit anatomical, physiological + pathological similarities to the human condition
• Discrimination = refers to the model’s ability to distinguish between different conditions, treatments, or interventions.
  –> High-discrimination models provide clear data that can be used to make informed decisions about the efficacy of interventions.
• validation criteria for induced animal models:
  o external valitidty: construct, face & predictive valitdity
  o reproducibility: induced condition should be consistently reproducible across different labs & in different cohorts, ensuring reliability of the model
  o specificity: model should specifically induce the condition of interest without causing unrelated symptoms or diseases
• validation criteria for spontanoeus models:
  o external valitidty: construct, face & predictive valitdity
  o genetic & phenotypic similarity: model should naturally develop the disease due to genetic or environmental factors that are analogous to those causing the disease in humans
  o model should exhibit a similar progression as seen in humans
  o clinical relevance: symptoms, disease course & complications should be relevant to human clinical presentations

Question 2

The size of our experiment is dictated by four elements: Variation, effect size, level of significance, and statistical power. Choose two of these elements, explain what they are, and how they influence how many animals we need to use.

Answer 2

• variation: the measure of how much the data points differ from each other or from the mean; indicates the degree of dispersion in the dataset.
  –> the bigger the variation the bigger the sample size
• effect size: the magnitude of the difference being studied
  –> the bigger the effect size, the smaller the sample size but bigger potential suffering
• level of significance: threshold for determining whether an observed effect is statistically significant, set at 5% & represents the probability of Type I error
• statistical power: probability of correctly rejecting the null hypothesis when it is false (probability to avoid a Type 2 error)

Question 3

Describe relevant factors and possible sources of bias, when planning an animal experiment, and possible actions to prevent them.

Answer 3

• selection bias = Assigning animals to treatment groups based on characteristics that may affect the outcome
  –> Prevention: randomization & blinding
• performance bias = treating animals differently based on their group assignment.
  –> Prevention: blinding & standardized protocols
• pilot studies help to account for variability & all types of biases: discover preventable issues before doing the proper study
• prevention via blocking

Question 4

When planning an experiment, what do we mean by “experimental unit?” Give two examples of experimental units (of different “size”).

Answer 4

smallest entity that receives the treatment/intervention & on which measurements or
observations are made to assess the treatment effects.
* 1. Individual Animal (Smaller Experimental Unit)
o Each animal can be considered an experimental unit
o In a drug efficacy study, each mouse is treated with different doses of the
drug, and their response is measured and compared within one subject.
* 2. Group of Animals (Larger Experimental Unit)
▪ a group of animals collectively is treated as a single experimental unit,
when treatments are applied to the group rather than individually.
▪ The response of the entire group is measured collectively, rather than
individual responses within the group.
▪ Mice are housed together in a single cage & the group is fed a specific
diet. The average weight gain of the mice in the cage is measured.

Question 5

Why do we decide on the number of animals we will be using in an experiment before we start; and how can we do that in a sensible way?

Answer 5

• Moral Obligation & legal compliace with the 3Rs
• power analysis
• pilot study

Question 6

ometimes when we plan experiments, we realize that there are benefits to using something other than a simple case-control study. Give an example of a more complex design, of when we might use it, and why.

Answer 6

• cross-over design:
  o we use animals as their own control
  o treatments are separated in time but are carried out on the same subject
  o in the middle of the experiment we “swap” conditions
• Factorial design
  o study the effects of two or more factors simultaneously to observe the interaction between these factors.
  o e.g.: experiment to investigate effects of 2 different diets (A + B) & 2 different exercise regimens (X + Y) on the weight gain.
  o Groups:
   Diet A with Exercise X & Diet A with Exercise Y
   Diet B with Exercise X & Diet B with Exercise Y
  o fewer animals are required to conducting separate experiments for each factor

Question 7

Sometimes when planning an experiment we may need outside help with our statistics. Give an example of when this might happen and of the elements of your study that you need to be able to present to, for example, a statistician.

Answer 7

• seek help when: complex experimental deisgn, advanced statistical methods, sample size calculation & data management & anylsis
• present study design, effect size, significance level, set power,

Question 8

How can pilot studies help us when planning our experiments? Can we use a pilot study to estimate an effect size?

Answer 8

• Pilot studies are small-scale preliminary studies conducted to refine and test the feasibility of the main study’s design
• test feasability, refine study design, train, estimate variability, sample size calculation, ensure safety/animal welfare
• considerations & limitations when using a pilot study to estimate effect size
  o small sample sizes = effect size estimates might not be precise.
  o conditions in the pilot study might not fully replicate those of the main study, which could affect the applicability of the effect size estimate.

Question 9

Explain the principles of GLP in experimental animal research.

Answer 9

Good laboratory practice:

• set of principles intended to ensure quality, integrity & reliability of laboratory studies
• comprehensive & consistent standard operating procedures (SOPs)
• ensures that experiments are conducted in a consistent & controlled manner, leading to reliable and reproducible results.
• Compliance is often required for the approval of new products, ensuring that the data submitted is credible and scientifically valid.

Question 10

Explain key issues to be reported when publishing experimental animal studies.

Answer 10

• give all information to be able to reproduce the results
  –> reproducibility crisis
• ARRIVE guidlines
• Study Design: Clearly describe the experimental design, including the type of study, allocation of animals to treatment groups & any blinding or randomization
• Animals: include species, strain, age, sex, weight, …
• Interventions and Procedures: describe interventions or treatments administered to the animals, including dosage, route of administration, frequency, & duration.
• Provide step-by-step details of experimental procedures, measurements, and assessments conducted on the animals.
• Outcomes: Report how outcomes were measured: tools, instruments& techniques
• Results: Report all relevant quantitative data, including means, standard deviations, and measures of statistical significance.
• Discussion: Discuss the strengths and limitations of the study, including potential sources of bias, confounding factors, and uncertainties.