Sat in lockdown in the midst of a global pandemic makes for an interesting time to be studying stress! So it seemed liked a great time to write a blog summarising what we know so far about stress as a risk factor for dementia, and specifically dementia caused by Alzheimer’s disease.
Stress when it’s time limited and manageable can be beneficial to us (who else needs an adrenaline hit to get that piece of work over the line!). But when stress levels get out of control, or run on for too long, it can be harmful to our health in many ways. There is a growing body of research investigating the links between cortisol, the stress hormone in humans, and Alzheimer’s disease.
What do we know from animal studies?
Studies have exposed mice to stress and looked at the effect this has on Alzheimer’s disease pathology (amyloid and tau). The studies use animals who are genetically modified to develop Alzheimer’s disease and we know when these animals typically develop the pathology. Increases in both amyloid and tau have been seen after exposure to stress compared to control animals (Cuadrado-Tejedor et al, 2012).
Scientists have also looked at the effect of Alzheimer’s disease on stress hormone levels in animals. Rodents with either amyloid pathology or genetic mutations for Alzheimer’s disease have higher levels of corticosterone (the rodent stress hormone) compared to control animals (Brureau et al, 2013).
These studies suggest that there is an interesting relationship between stress, corticosterone and Alzheimer’s disease pathology in animals, but we still don’t understand which way the relationship works- is stress having an impact on the disease process, is the disease process increasing stress hormone levels, is it both or is this association actually a by-product of some as yet unknown factor?
What do we know about cortisol and Alzheimer’s disease dementia and preclinical AD?
When researchers have studied cortisol levels of people with Alzheimer’s disease they discover that higher cortisol levels are associated with a more rapid decline in their memory and thinking skills (Csernansky et al, 2006). High cortisol levels can also predict worsening in the Mini-Mental State Examination (MMSE) over time (Huang et al, 2009).
The Australian Imaging, Biomarkers and Lifestyle Flagship Study of Ageing (AIBL) found that increased plasma cortisol in amyloid beta positive cognitively normal older adults (a group we might classify as having pre-clinical Alzheimer’s disease- i.e. the disease is present in the form of amyloid beta but there are no symptoms) was associated with a faster cognitive decline (Pietrzak et al, 2017). A similar finding was recently published from the Alzheimer’s Disease NeuroImaging Initiative (ADNI) cohort with high levels of cortisol associated with clinical progression from cognitively normal in those who are amyloid beta positive. Interestingly in this study high reserve (calculated using measures of intracranial volume on brain scans and lifetime experiences) was able to reduce the associated progression risk (Udeh-Momoh et al, 2019).
These studies suggest that higher cortisol is associated with faster progression of symptoms in both preclinical and clinical populations- although we do not yet understand the mechanisms at play here. We also do not know from these if the cortisol levels were higher pre-disease or if the disease process has interacting with the stress system in some way to increase the levels.
What do we know about cortisol and cognitively healthy older adults?
Given the above it is important to take a step back in the disease process and look at a population who at baseline are not known to have any symptoms. When we do this we also see similar trends in older adults who are free from symptoms of cognitive impairment. In one community-based study older adults (mean age of 78 years) provided saliva samples of cortisol 3 times a day once a year for 3 years. The study found that those with a high evening cortisol (taken around 11pm) had a steep decline in delayed paragraph recall scores over time (Li et al, 2006) . This memory task was used as a measurement of delayed verbal memory and was the primary outcome in the study. Another study used overnight urine collection to test for cortisol levels in a similarly aged population(538 adults aged 70-79 at baseline) and found that those with higher cortisol had an increased risk of developing cognitive impairment over the 7 year follow up compared to those with the lowest levels of cortisol (Karlamanga et al, 2005).
One limitation of both of these studies is that whilst we know they were symptom free at baseline, we do not know if they had any pathological beginnings of disease in their brain making them ‘preclinical’ rather than ‘cognitively healthy’.
Can we intervene?
Hopefully! Whilst this might be an interesting research area, it is also important for us to consider the bigger picture of whether further investigations in this area could lead to developments for interventions.
Drugs have been trialled and continue to be developed to target this specifically in Alzheimer’s disease (Marek et al, 2014 & Webster et al, 2017). With Alzheimer’s disease proving difficult for pharmaceutical compounds to successfully treat or prevent, the broadening out of potential targets can only be beneficial to the field.
There is also likely to be a whole host of non-drug interventions that we could trial with the aim of reducing the impact of stress on the body and investigating if this has any impact on disease or symptom progress. Things such as yoga (Brenes et al, 2019), exercise (Tortosa-Martínez et al, 2018), sleep and diet (Pistallato et al, 2016) could all have merit for further exploration. As with many areas of research in dementia, it is likely that this will be one of many areas to understand for an extra piece of the puzzle and additional treatment avenues.
What else do we need to know?
There is a huge amount we still do not understand about how stress affects the risk, onset and progression of Alzheimer’s disease dementia. We also do not know how it impacts on other diseases that lead to dementia such as vascular or frontotemporal dementia. We are still developing intervention strategies as we understand more about this.
In my PhD I am focusing on understanding when these changes in stress hormones might begin and if there is a targetable group who would be most likely to benefit from therapeutic trials to change cortisol levels with the aim of changing Alzheimer’s disease risk.
Sarah Gregory is a PhD Student & Study Coordinator at the University of Edinburgh. Her PhD investigates the HPA axis in midlife and it’s association with dementia, and in addition to her studies she words as a study coordinator on the PREVENT Study. Previously she worked in a research team in London coordinating multiple clinical trials of investigational medicinal products of novel dementia treatments, and even once served foot to Charles and Camilla.
You can follow Sarah on Twitter Follow @GregorySarah