SENSITIVITY OF AN ALZHEIMER’S DISEASE COOPERATIVE STUDY: PRECLINICAL ALZHEIMER’S COGNITIVE COMPOSITE (PACC) BASED ON COMPUTERIZED TESTS TO AMYLOID LOAD IN PRECLINICAL AD

Abstract

Secondary prevention clinical trials of amyloid (AB) lowering drugs use the ADCS-PACC, a composite of verbal list learning, paragraph recall, executive function and the MMSE] as their primary endpoint. This study simulated the effect of lowering AB levels on estimates of cognition and clinical disease progression using data from participants who satisfy enrollment criteria for the A4 or EARLY studies but who had varying levels of amyloid. we compared the effect of lowering amyloid on PACC composite scores consisting of computerized and paper and pencil tests. Cognitively normal (CN) adults with data for 36 months of follow-up enrolled in the Australian Imaging Biomarkers and Lifestyle (AIBL) study and defined as having abnormally high AB from PET scanning were classified into two subgroups, AB+ (SUVR>1.4 and <1.9, n = 138) and AB++ (SUVR = />1.9, n = 89]. AB groups were compared on rates of disease clinical progression using X2 and on rates of cognitive decline using ANCOVA. PAAC scores consisting of comnputerized tests from the Cogstate battery were compared to that consisting of verbal list learning, logical memory, coding and MMSE. Statistical models were adjusted for clinical or demographic factors that also differed between the groups. ANCOVA models showed that rates of disease progression were lower in the AB+ [10%] than in AB++ group [5.6 %; OR 5.1 95%CI 3.4 -6.7]. ANCOVA showed that the rate of decline on the conventional ADCS-PACC was lower in the AB+ group than in the AB++ group: with [d = 0.89] or without [d = 0.65] progressors included in analyses. Using the Cogstate tests to measure episodic memory and executive function difference in rate of decline between AB+ and AB++ groups was equivalent with [d = 0.86] or without [d = 0.66] progressors included in analyses. This data provides a basis for the extent to which a change in amyloid levels, measured by amyloid plaque burden, could manifest in terms of change in cognition or clinical disease progression. Is also shows computerized measures can be used in PACC scores.