Advertisement

PATHOLOGY IN THE EARLY ALZHEIMER’S PRODROME: SELF AND INFORMANT COGNITIVE CONCERNS SHOW DIFFERENT PATTERNS OF ASSOCIATION WITH TAU DEPOSITION ON [18F]FLORTAUCIPIR PET IN AMYLOID-POSITIVE OLDER ADULTS

      Background

      The presence of subjective cognitive decline, characterized by self-perception of significant decline in cognition in the absence of measurable psychometric deficits, is thought to be a risk factor for progression to Alzheimer’s disease. Similar concerns by a knowledgeable informant confers additional risk. We compared the associations between subjective and informant based concerns and tau deposition measured using [18F]Flortaucipir PET in cognitively normal or mildly symptomatic older adults positive for amyloid deposition.

      Methods

      31 participants from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) were included, including 13 cognitively normal older adults without concerns (CN), 4 older adults with significant memory concerns (SMC), and 14 older adults with early mild cognitive impairment (EMCI). Amyloid positivity was defined as mean cortical [18F]Florbetapir SUVR > 1.10. Pre-processed [18F]Flortaucipir scans were downloaded from LONI and processed using standard techniques. Briefly, scans were co-registered to the closest timepoint structural MRI, normalized to standard space, and intensity normalized to mean cerebellar crus uptake to create SUVR images. Regional [18F]Flortaucipir from the inferior temporal lobe, frontal lobe, parietal lobe, and global cortex were extracted from subject-specific regions of interest generated from Freesurfer. Partial Pearson correlation and linear regression models were used to evaluate the relationship of the ECog with mean [18F]Flortaucipir SUVR in the target regions and on a voxel-wise level (p<0.001 (uncorrected); k=50 voxels), covaried for age and sex.

      Results

      Significant associations between both self and informant ECog scores and [18F]Flortaucipir SUVR were observed (Figure 1). Notably, the self-based memory concerns were associated with tau load predominantly in the frontal cortex and medial temporal regions, while the informant-based concerns were more strongly associated with precuneus and lateral parietal tau load (Figure 1C). However, informant-based concerns were associated with more widespread regions than self-based concerns on voxel-wise analysis (Figure 1B).

      Conclusions

      The association between subjective concerns and tau deposition suggests an important role of tau pathology in the early stages of disease and that involvement of different brain regions underlie the phenomena detected by the ECog scale. Future longitudinal studies will help to elucidate the timing and source of cognitive concern presentation relative to disease and pathological progression.
      Figure thumbnail fx1