Memory Resilience to Alzheimer's Genetic Risk: Sex Effects in Predictor Profiles

Kirstie L. McDermott, G. Peggy McFall, Shea J. Andrews, Kaarin J. Anstey, Roger A. Dixon*

*Corresponding author for this work

    Research output: Contribution to journalArticlepeer-review

    34 Citations (Scopus)

    Abstract

    Objectives Apolipoprotein E (APOE) ϵ 4 and Clusterin (CLU) C alleles are risk factors for Alzheimer's disease (AD) and episodic memory (EM) decline. Memory resilience occurs when genetically at-risk adults perform at high and sustained levels. We investigated whether (a) memory resilience to AD genetic risk is predicted by biological and other risk markers and (b) the prediction profiles vary by sex and AD risk variant. Method Using a longitudinal sample of nondemented adults (n = 642, aged 53-95) we focused on memory resilience (over 9 years) to 2 AD risk variants (APOE, CLU). Growth mixture models classified resilience. Random forest analysis, stratified by sex, tested the predictive importance of 22 nongenetic risk factors from 5 domains (n = 24-112). Results For both sexes, younger age, higher education, stronger grip, and everyday novel cognitive activity predicted memory resilience. For women, 9 factors from functional, health, mobility, and lifestyle domains were also predictive. For men, only fewer depressive symptoms was an additional important predictor. The prediction profiles were similar for APOE and CLU. Discussion Although several factors predicted resilience in both sexes, a greater number applied only to women. Sex-specific mechanisms and intervention targets are implied.

    Original languageEnglish
    Pages (from-to)937-946
    Number of pages10
    JournalJournals of Gerontology - Series B Psychological Sciences and Social Sciences
    Volume72
    Issue number6
    DOIs
    Publication statusPublished - 1 Nov 2017

    Fingerprint

    Dive into the research topics of 'Memory Resilience to Alzheimer's Genetic Risk: Sex Effects in Predictor Profiles'. Together they form a unique fingerprint.

    Cite this