AB+ CLINICALLY NORMAL PARTICIPANTS WITH ELEVATED TAU SHOW GREATEST DECLINE IN THE PRECLINICAL ALZHEIMER’S DISEASE COGNITIVE COMPOSITE

Aggregation of Aβ and Tau are thought to be critical components of the Alzheimer’s disease cascade. Using PET imaging to assess these hallmark pathologies, we sought to determine their joint contribution to cognitive decline during the preclinical stage of AD. Two hundred and seventy seven clinically normal participants from the Harvard Aging Brain Study have completed baseline PIB PET scanning and at least one follow up annual neuropsychological testing session (age=74±6, mean follow up=3.3±1.1 years). A subset of 130 have underwent Tau PET imaging with T807/AV-1451 (age=73±6, mean follow up=3.7±0.9). Using linear mixed models, we examined the association between Aβ status and inferior temporal Tau with longitudinal decline on the Preclinical Alzheimer’s disease Cognitive Composite (PACC). The PACC is comprised of the Logical Memory delayed recall, Free and Cued Selective Reminding Test (FCSRT) total cued recall score, the MMSE total score, and the Digit Symbol total score. Age, sex, and education were covaried. In the full sample, Aβ+ CNs showed significantly more decline in PACC compared to Aβ- CNs (beta=-0.08, p-value=0.03), and this difference was significant when examining individual components of the PACC separately (p-values<0.04). When both Aβ status and inferior temporal Tau were modeled as simultaneous predictors in the subset of 130 CN, Aβ status was not significant (beta=-0.006, p=0.82) whereas higher Tau was associated with decline (beta=-0.47, p=0.0007). Inclusion of an interaction term between Aβ and inferior temporal Tau revealed a significant effect (p=0.02), such that the association between inferior temporal Tau and decline in PACC was only present among Aβ+ CN. Among Aβ+ CN, inferior temporal Tau was significantly associated with decline in individual PACC components (p-values<0.05) with the exception of the FCSRT (p=0.12). Inferior temporal Tau was not associated with decline in any individual test among Aβ- CN (p-values>0.63). Greater risk of short-term decline in Aβ+ CN that also have elevated Tau is consistent with the current conceptualization of preclinical AD. These high risk individuals may be ideal candidates for prevention trials using cognitive endpoints. However, whether efficacy of anti-amyloid treatment will be optimal in Aβ+ CN with either elevated or low Tau is unknown.