Date Published: March 24, 2018
Author(s): Cecily G. Swinford, Shannon L. Risacher, Arnaud Charil, Adam J. Schwarz, Andrew J. Saykin.
•Both self– and informant memory concerns are associated with tau aggregation in at-risk adults.•Self-concerns are more associated with frontal tau compared to a parietal pattern for informant concerns.•This pattern is strongest in amyloid-positive participants.
Alzheimer’s disease (AD), the leading cause of neurodegenerative dementia associated with aging, affects over 5 million adults in the United States and is predicted to increase to 16 million affected by 2050 . There are presently no approved pharmacological treatments that can stop the progression of AD. Treatment is likely to be most effective during the preclinical or early prodromal stages of AD, before substantial permanent neurodegenerative and cognitive damage has occurred. Therefore, there has been considerable recent interest in measures to identify older adults at highest risk for progression to AD who may benefit most from early intervention , .
In this study, we found that self– and informant memory concerns are both correlated with tau aggregation in at-risk adults, but that the overall spatial patterns of associations differ between the two measures of subjective memory concern. We found that self– and informant ECog memory scores are only mildly correlated with one another in at-risk adults and are not correlated with one another in the subset of amyloid-positive adults, suggesting that the two measures of subjective memory complaints may be somewhat independent and provide complementary information. The results from our ROI-based analyses suggest that self-based memory complaints are most strongly associated with tau aggregation in the frontal lobes, whereas informant memory complaints are most strongly associated with tau aggregation in the parietal lobes. These differences in patterns of correlation were enhanced when only amyloid-positive participants were included, and significant source-by-region interactions were found in both the full analysis and the subanalysis of amyloid-positive individuals. There were no correlations found when only amyloid-negative participants were included, suggesting that the patterns of association found are specifically present in adults at higher risk for developing AD, which supports the idea that these subjective memory tools may be useful for optimizing screening techniques in the population of at-risk adults. In addition to our regional analyses, we evaluated the associations with voxelwise analyses to lessen the bias imposed by our predetermined ROIs. Although these analyses revealed regions of overlap between the correlations of tau aggregation with self– and informant memory concerns, they resulted in the same general patterns found with the ROI-based analyses. Specifically, both self– and informant memory concerns were associated with tau aggregation in the frontal and temporal lobes when all participants were included, while the informant memory concerns were also associated with tau aggregation in posterior brain regions, including the parietal and occipital lobes. When only amyloid-positive participants were included, the self-based memory concerns were significantly associated with tau aggregation in bilateral frontal lobes and the right temporal lobe, while the informant memory concerns were also correlated with tau burden in the left parietal lobe and bilateral occipital lobes.