Premium
Relationship between genetic risk for Alzheimer's, cognition and neuropsychiatric symptoms: A case study of DNA sampling and analysis through digital platform cohort studies
Author(s) -
Creese Byron,
Brooker Helen,
Aarsland Dag,
Corbett Anne,
Ballard Clive,
Ismail Zahinoor
Publication year - 2020
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1002/alz.040237
Subject(s) - dementia , cohort , cognition , informed consent , cohort study , alzheimer's disease , clinical psychology , psychology , medicine , disease , psychiatry , pathology , alternative medicine
Background With the evolution of technologies to extract DNA and undertake analysis from saliva samples, it is now feasible to collect postal samples for genetic analysis as part of cohort studies run on digital platforms. Ethics and engagement will be discussed and Mild Behavioral Impairment (MBI) is presented as an example of where this approach has delivered a substantial and cost‐effective study. MBI, a late‐life neuropsychiatric syndrome, is associated with faster progression to dementia and Aβ deposition even in healthy adults. MBI screening may enrich samples with individuals at risk for dementia, with benefits to clinical studies and possibly trials. Method 25,000 participants from the PROTECT digital platform were invited to provide a saliva sample by post. Ethics approval was underpinned by an on‐line consent process, genotypes were not disclosed to participants. Polygenic scores (PRS) for Alzheimer’s disease (AD) were calculated and split by tertile (representing low, medium and high AD genetic risk). Data were analysed as a whole sample then stratified by the presence of MBI. Result 87% of PROTECT participants agreed to provide genetic samples by post. 10,000 of these have been genotyped. 75% of the sample were women and the mean age was 62 (range: 50‐100). At the time of data freeze, genetic, MBI and cognitive data were available for 3,126. AD genetic risk was associated with a lower cognitive score (F(2,3119)=3.93, p=0.02; mean difference between low and high genetic risk: ‐15, p=0.02, Cohen’s d=0.13). In stratified analysis, genetic risk for AD was associated with worse cognition but only in the MBI group (MBI: F(2,1746)=4.95, p=0.007; no MBI: F(2,1366)=0.72, p=0.49). There was a significant difference between the high and low genetic risk groups (mean difference: ‐0.22, p=0.005); the effect size was stronger than in the whole sample analysis (Cohen’s d increase from 0.13 to 0.19). Conclusion High degrees of engagement can be achieved to obtain DNA samples from participants in online cohort studies. With respect to MBI, these findings demonstrate that neuropsychiatric symptoms may modify the relationship between genetic risk for AD and cognitive impairment. MBI screening may represent a useful sample enrichment strategy for clinical studies and trials.