No evidence for association of Alzheimer’s disease risk loci with the rate of cognitive decline among elderly with type 2 diabetes

Abstract

Multiple risk loci for late-onset Alzheimer's disease (LOAD) have been identified in recent years. Type 2 diabetes (T2D) is a well-established risk factor for cognitive decline and dementia. In the Israel Diabetes and Cognitive Decline (IDCD) cohort, we investigated the association of LOAD loci with rate of cognitive decline. Participants were Jewish elderly (>65 years old) with T2D, cognitively normal at baseline, who were followed up every 18 months for a period of 54 months. Assuming an additive genetic model, mixed regression models (adjusted for age, sex, education and ancestry) examined the associations of each of the variants with global cognitive decline. We also calculated a polygenic risk score (PRS), aggregating effects of all variants. Last, we examined the association of LOAD loci and the PRS with hippocampal volume (N=190) measured at the IDCD baseline, as determined by brain magnetic resonance imaging (MRI). Twenty-two SNPs, the APOE E4 allele and the PRS were examined in 944 participants (average age at baseline: 72.5 years). Following correction for multiple testing, none of the variants was associated with change in global cognitive functioning. Results were similar for associations with specific cognitive domains of episodic memory, attention/working memory, executive functions and language or with hippocampal volume. Adjusting also for diabetes-related and cardiovascular risk factors did not substantially change the results. Our results do not support contribution of LOAD loci to cognitive decline nor to hippocampal volume in a population of older adults with T2D. This suggests that other genes and mechanisms may underlie the consistently higher risk of dementia in T2D.