Using CANTAB to investigate age-related cognitive changes associated with Alzheimer's disease in Down syndrome

PhD researcher, Rosalyn Hithersay, spoke to us about her recent publication on which cognitive tests are most sensitive to the early stages of dementia in Down syndrome.

Can you tell us more about your research group?

The London Down Syndrome Consortium (LonDownS) are a multi-disciplinary group of researchers and clinicians investigating Alzheimer’s disease in people with Down syndrome: from the genetic and cellular levels, through to the cognitive changes seen in adults as they get older.

For our recent publication, we collaborated with Nicholas Firth at UCL’s Progression of Neurodegenerative Diseases (POND) group to explore, for the first time, the sequence of decline in Alzheimer’s disease in Down syndrome, using an Event-Based Model.

What is the rationale behind your study?

Dementia is an exceptionally common part of growing old with Down syndrome because by their mid-30s, almost everyone with Down syndrome will have developed plaques of amyloid protein that we associate with Alzheimer’s disease (Mann & Esiri, 1989). Dementia onset in Down syndrome is typically young, with an average age of diagnosis of just 55 years (Sinai et al., 2018). Yet some individuals survive well into their 60s without showing clear signs of decline. Our group are keen to understand the factors that explain this variability. Further, despite the known links between Down syndrome and Alzheimer’s disease, questions remain concerning the sequence of decline due to Alzheimer’s disease in this population.

Having previously validated a substantial battery of cognitive tasks, including CANTAB measures and more traditional table-top tasks (Startin et al., 2016), in our recent paper, 'Ageing related cognitive changes associated with Alzheimer's disease in Down syndrome'¹, we used data from these cognitive measures and associated informant report measures, to explore the sequence of cognitive decline in adults with Down syndrome.

Which methods did you use?

We employed an Event-Based Model (EBM) to sequence ageing-related change in a selection of 10 cognitive tasks and 3 informant measures from the LonDownS battery.

We included data from 283 LonDownS participants: 119 were aged 35 years or younger. These younger adults are considered a control group, who should not yet be showing substantial Alzheimer’s disease neuropathology. Those aged over 35 (N=164) can be expected to be showing amyloid deposition and are considered our ‘disease’ group, with many likely to be showing signs of deterioration in their abilities.

The EBM considered scores from each of our tasks as a ‘biomarker’ for dementia. Using data from younger and older adults separately, the model is able to pick out underlying patterns in the test scores that can be associated with dementia, while controlling for pre-decline levels of intellectual disability. A maximum likelihood event sequence (with corresponding sampling uncertainty) was obtained, showing the order that different test’s scores change as dementia progresses.

What are your key findings?

Our study found that the tasks showing the earliest changes in the staging model were the first trial memory score from the CANTAB Paired Associates Learning task and the NEPSY-II Visuomotor-Precision task, suggesting that measures of sustained attention and associative memory are most sensitive to dementia-related decline in the early stages. In the middle stages, we begin to see decline in measures of rule-learning / switching using CANTAB Intra-Extra Dimensional Set Shift task, and in reaction time using the CANTAB Simple Reaction Time task. Informant measures then provided further detail of decline at the latest stages of the disease.

What are the implications of your study?

Measuring cognition in individuals with Down syndrome can be challenging, as most people in this population have some degree of intellectual disability. Our results show that we can not only measure varied aspects of cognitive function in this population using tasks from the CANTAB battery, but that some of these tasks are sensitive enough to track decline even before the onset of dementia.

This is the first time that an EBM approach has been applied to cognitive data from the Down syndrome population, allowing us to look at Alzheimer’s disease progression in a new way. Our results will be helpful for designing future clinical trials, and providing more detailed prognoses in the clinic.

Why did you choose CANTAB?

Our research examines individual differences in abilities in a challenging population. We require tasks that are easily administered, yet capable of producing reliable and replicable results both across our individuals, and in the same individuals over time.

CANTAB provide tasks covering a variety of cognitive domains, often with varied settings to accommodate the needs of a given clinical population. Many of the tasks are non-verbal, removing confounds of language impairments, which are common in people with Down syndrome. This also means that we can collaborate widely with other groups, in different countries, which is important for larger studies.

What are the next steps for your research?

We are tracking the LonDownS participants longitudinally. Our future work will provide data from different time points, to re-assess the validity of our events-based model.

Our work also includes genetic and blood biomarker data from the same individuals that have provided extensive cognitive test scores. Future work will explore how these different factors are related, and how they interact in the progression of dementia in this population.

Find out more

¹Study Authors: Nicholas C. Firth, Carla M. Startin, Rosalyn Hithersay, Sarah Hamburg, Peter A. Wijeratne, Kin Y. Mok, John Hardy, Daniel C. Alexander, and André Strydom.

Rosalyn Hithersay, PhD Student at University College London working with the London Down Syndrome Consortium