In the last three decades significant advances have been made in our understanding of what happens in the brain of someone with dementia. Animal studies have made a large contribution to our knowledge of the brain diseases which lead to dementia, but this knowledge has failed to translate into effective treatments. In this blog I will be discussing whether a reliance on animal models of disease has delayed progress in dementia research.
Focussing specifically on Alzheimer’s disease as it is the most common brain disease which leads to dementia, the failure rate in drug development has been reported to be as high as 99%. Why have preclinical studies using Alzheimer’s disease mouse models failed to translate into successful clinical trials? We first need to consider why we might want to use an animal model in the first place and what a good animal model actually looks like. Preclinical animal studies are a requirement for drug development providing information on whether a drug will be safe and efficacious before human clinical trials are conducted. A good animal model of disease should show the typical characteristic features of the disease, which includes pathological and behavioural changes. For example, we should be able to observe the characteristic features of amyloid plaques and neurofibrillary tangles, as well as degeneration of synapses and neurons. These pathological changes should be associated with cognitive-behavioural impairments which develop as the disease progresses.
We have established that a good animal model needs to show specific features of a disease, so how well do mouse models of Alzheimer’s disease reflect the pathological changes associated with the disease? Preclinical research on Alzheimer’s disease heavily relies on mouse models whereby the disease characteristics can be induced via genetic modification; they are known as transgenic mice. Many of these models are based on the genetic mutations observed in familial Alzheimer’s disease which is hereditary and has an earlier onset compared to sporadic late onset Alzheimer’s disease which accounts for around 97% of all cases. There are no established, well characterised mouse models of sporadic Alzheimer’s disease, which has led to a reliance on transgenic models. However, these models are limited in that none fully mimic Alzheimer’s disease in humans. For example, a transgenic model will only display some of the associated characteristic pathologies, which fails to reflect the complexity of the disease. Although we don’t know exactly how Alzheimer’s disease is initiated, it is likely to manifest due to genetic, lifestyle, and environmental factors, with age being the largest risk factor of all. We therefore have to question whether any genetically modified animal model with a relatively short lifespan can mimic a sporadic disease which progresses for decades in humans before any clinical symptoms are apparent.
Mouse models of Alzheimer’s disease can only reflect limited pathological aspects of the disease, but are they any better at reproducing the cognitive-behavioural changes which can include impairment in memory and visuo-spatial skills, for example. These changes can be assessed in humans using various different neuropsychological tests, which provides an indication of cognitive function as an outcome measure to monitor disease progression. Cognitive-behavioural tasks have been developed for mice, but if we consider the main cognitive feature of Alzheimer’s disease as being a decline in memory, how can we ask a mouse to, for example, recall a list of words? Obviously, we can’t, so these tasks have been adapted to assess memory function as an indirect measure of their behaviour. For example, mice naturally seek to explore novel objects so we can compare how much time they spend exploring a novel object compared to a familiar object which they’ve seen before. If they remember the familiar object, they will spend more time exploring the novel object. If they do not remember the familiar object, they will treat both objects as novel. Considering these differences in how cognitive outcome measures are assessed, we must question to what extent cognitive function in these animal models are comparable to humans, and whether cognitive impairment reflects the decline associated with Alzheimer’s disease in humans.
Having established that mouse models of Alzheimer’s disease do not show all the pathological and behavioural features of the disease, it is important to recognise that animal studies can often be poorly designed, with a lack of standardisation of outcome measures, and animal housing and husbandry. These issues are not unique to preclinical Alzheimer’s disease studies, but they do reflect wider challenges on translatability of animal to human studies.
So what is the solution? Although progress in finding effective treatments for dementia-related brain diseases like Alzheimer’s may have been delayed by the difficulties in developing an animal model which fully reflects the nature of the disease in humans, the challenge remains to establish the mechanisms of how the disease is initiated. However, mouse models of Alzheimer’s disease which only reflect certain aspects of the disease have still advanced our knowledge on the various mechanisms and pathologies which contribute to the progression of the disease. Understanding how genetic, lifestyle, and environmental factors interact to trigger the onset of the disease will be an important challenge to overcome and could unlock novel areas for developing therapies. In the meantime, the quality and reproducibility of animal studies can be enhanced by improving the design and reporting of animal studies using the ARRIVE (Animals in Research: Reporting In Vivo Experiments) and PREPARE (Planning Research and Experimental Procedures on Animals: Recommendations for Excellence) guidelines. Pre-registration of animal experiments should also be encouraged to improve transparency. Finally, when translating a disease which results in distinct cognitive impairments, the cognitive outcome measures, which are so heavily relied on to tell us whether a disease model has face validity and the efficacy of an intervention, need to be back-translated into humans so we can be confident that when we are measuring a cognitive function in an animal model, it reflects the same processes which occur if measured in an analogous human task.
Dr Kamar Ameen-Ali is a Research Associate at University of Glasgow, exploring how neuroinflammation following traumatic brain injury contributes to the progression of neurodegenerative diseases that lead to dementia. Having first pursued a career as an NHS Psychologist, Kamar went back to University in Durham to look at rodent behavioural tasks to completed her PhD, and then worked as a regional Programme Manager for NC3Rs. Kamar brings a wealth of experience and writes on a range of topics from her time in the NHS, working for a Research Funder and from her work and life in the lab.