Gene-dose effects of APOE ε4 on age-dependent increases in Peak Width of Skeletonised Mean Diffusivity during midlife

• Published in: Cerebral circulation - cognition and behavior 2024, Vol.6, p.100293, Article 100293
• Main Authors: Mak, Elijah, Dounavi, Maria-Eleni, Low, Audrey, Terrera, Graciela Muniz, Malhotra, Paresh, Koychev, Ivan, Mackay, Clare, Lawlor, Brian, Naci, Lorina, Wells, Katie, Ritchie, Craig, Ritchie, Karen, Su, Li, Gispert, Juan, O'Brien, John
• Format: Article
• Language: English
• ISSN: 2666-2450; 2666-2450
• DOI: 10.1016/j.cccb.2024.100293

The mechanistic associations between small vessel disease (SVD) and dementia are still poorly understood. The APOE ε4 allele, recognised as the strongest genetic risk factor for Alzheimer's disease, has been previously implicated in SVD, although it remains unclear whether this association is gene-dose dependent. An emerging neuroimaging biomarker of SVD is Peak Width of Skeletonised Mean Diffusivity (PSMD), obtained from histogram analyses of diffusion weighted imaging (DWI) datasets. Here, we investigated the relationship between APOE ε4 gene dose and PSMD, as a surrogate marker of SVD, in a group of cognitively normal middle-aged adults. The study included data from 1954 asymptomatic middle-aged adults from the ALFA (ALzheimer and FAmilies) and PREVENT-Dementia cohorts (See Table 1 for sample characteristics). PSMD was calculated from the DWI datasets using a publicly available script, and harmonised using COMBAT to account for site-related differences. Using non-parametric permutation models, our primary analyses focused on the (a) comparison of group differences (APOE ε4 heterozygotes vs homozygotes vs non-carriers) in PSMD, adjusting for age, sex, years of formal education, and sites; and (b) potential interactions between APOE ε4 gene dose and age on PSMD values. Marginal predictions were used to estimate the earliest age at which differences might emerge between the APOE ε4 groups and non-carriers. There were no significant differences in PSMD values across the non-carriers (n=1,197), heterozygous carriers (n=659), and homozygous APOE ε4 carriers (n=98) (p = 0.6; Figure 1). However, there was a statistically significant interaction between APOE ε4 gene dose and age on PSMD. Specifically, homozygous APOE ε4 carriers exhibited a steeper increase in PSMD with age compared to non-carriers and heterozygous carriers (T = 4.7, p