Comprehensive risk prediction models and presymptomatic biomarkers for dementia

Through leveraging extensive longitudinal data and biospecimens already obtained from a large representative cohort of older Australians, this research will enable better characterisation of the epidemiology of dementia in Australia and the identification of novel biomarkers of risk.
ASPREE (ASPirin in Reducing Events in the Elderly) is a randomised placebo-controlled trial to determine the role of aspirin in the primary prevention of dementia and disability-free survival. 16,703 initially healthy Australians aged 70+ years were screened for cognitive impairment at study entry and followed for a mean of 5 years with regular comprehensive cognitive assessment. By study end (2017), 1000 individuals are estimated to develop dementia, and 2000 mild cognitive impairment. The wealth of information available on such a large cohort offers an unprecedented opportunity to generate new knowledge about modifiable risk factors for dementia (Alzheimer’s disease and related dementias), and protective factors for cognitive health and resilience. The availability of blood samples at baseline and follow-up (>12,000) will enable the discovery of novel preclinical biomarkers.
The principal aims of this research program will be:
(1) To build comprehensive models of dementia risk that incorporate environmental, social, genetic and biological factors. This information is essential to generate more precise predictive algorithms for dementia risk. The accurate identification of modifiable risk factors is crucial to the development of the most approriate interventions to delay onset or even prevent dementia.
(2) To identify a unique epigenome-wide DNA methylation signature to accurately differentiate individuals at study entry who go on to develop dementia for those who remain dementia free. The discovery of clinically relevant biomarkers will enable future development of earlier, presymptomatic diagnostic tests, thus allowing earlier interventions or treatments.