Premium
P3‐019: An Alzheimer's genetic risk composite, but not ApoE, intensifies diabetes‐related neurocognitive slowing in nondemented older adults
Author(s) -
McFall G. Peggy,
Sapkota Shraddha,
Wiebe Sandra A.,
Anstey Kaarin J.,
Dixon Roger A.
Publication year - 2015
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1016/j.jalz.2015.06.885
Subject(s) - neurocognitive , apolipoprotein e , type 2 diabetes , medicine , psychology , cognitive decline , disease , oncology , gerontology , dementia , diabetes mellitus , demography , cognition , psychiatry , endocrinology , sociology
Background:Risk factors (and their synergistic interactions) associated with Alzheimer’s disease (AD) may predict normal or preclinical deficits and decline. Although ApoE (rs429358, rs7412) is the gene most consistently linked with AD risk, genome-wide association studies have identified others, including CLU (rs11136000), CR1 (rs6656401), and PICALM (rs541458). Type 2 diabetes (T2D) is a risk factor for AD and for increased cognitive deficits in nondemented older adults. We examined if the effect of diabetes on neurocognitive speed performance (level) and longitudinal change was intensified by (a) genetic risk from each of the four variants independently or (b) an AD Genetic Risk Composite (AGRC) representing combined risk from all four variants. Methods: This longitudinal design included non-demented older adults (n1⁄4591, baseline M age1⁄469, age range 53–91, 68% women, 8% with T2D) followed over 9 years. Saliva was processed with standard procedures from Oragene-DNA Genotek. Genotyping was carried out using a PCR-RFLP strategy. The AGRC was created by summing allelic risk across the four specified genotypes: 01⁄4no risk, 11⁄4moderate risk, 21⁄4full risk and then grouped into low and high risk using median split (Mdn1⁄43.0). Statistical analyses included latent growth modeling testing independent and interactive effects on level (centering age1⁄475) and change using a confirmed neurocognitive speed latent variable consisting of choice reaction time, sentence verification, and lexical decision measures. Results:First, adults with T2D exhibited slower speed performance at age 75 than adults without T2D (b1⁄4.494, p1⁄4.007). Second, none of the genetic risk variants showed independent effects on speed performance or change. Third, interaction analyses (e.g., T2D x ApoE) showed no magnification of speed decrements. Fourth, intensification interaction analyses (T2D x AGRC) showed that adults in the high risk AGRC group with T2D exhibited significantly greater 9-year decline in speed (b1⁄4.048, p1⁄4.004). Finally, education and pulse pressure where significant covariates but did not alter the observed effects. Conclusions: Independently, as expected, diabetes was associated with cognitive slowing in nondemented aging. Only the AD Genetic Risk Composite, not ApoE or other variants, intensified the effects of T2D on neurocognitive speed in the form of exacerbated slowing over 9 years.