APOE‐ɛ4 Moderates the Association between HDL‐c and White Matter Hyperintensity Volume

TL;DR

Higher HDL cholesterol is linked to less brain damage in people with the APOE-ɛ4 gene variant, but not in those without it.

Contribution

This study shows that APOE-ɛ4 modifies the relationship between HDL cholesterol and white matter hyperintensity volume.

Findings

APOE-ɛ4 carriers showed a significant HDL-c×APOE-ɛ4 interaction on WMH volume.

Higher HDL-c levels were associated with lower WMH burden only in APOE-ɛ4 carriers.

No significant interactions were found between APOE-ɛ4 and LDL-c or triglycerides on WMH volume.

Abstract

Homozygous APOE‐ɛ4 carriers have exhibited heightened white matter hyperintensity (WMH) burden based on severity ratings (Rojas et al., 2018). A higher high‐density lipoprotein cholesterol (HDL‐c) to low‐density lipoprotein cholesterol (LDL‐c) ratio is associated with less severe WMHs (Wei et al., 2023). Additionally, higher concentrations of APOE protein are linked to higher cholesterol efflux in Alzheimer's Disease cohorts (Yassine et al., 2016). Despite the understanding of APOE as a lipid carrier (F. Yin, 2021), its mechanistic role in modulating dementia risk is still evolving. Here, we examine whether APOE‐ɛ4 positivity modifies the relationship between blood cholesterol levels and WMH volume in a multi‐ethnoracial cohort.

We examined 1645 cognitively unimpaired (CU) individuals from the Health and Aging Brain Study‐Health Disparities cohort (65.96% female, 25.6% APOE‐ɛ4+, aged 50‐90) (Table 1). Participants underwent a MR scanning (Siemens 3T Skyra or Vida), which included a T2 FLAIR image. We calculated WMH volume using SPM's lesion growth algorithm (LGA) and ran robust linear regressions to test for an interaction between log‐transformed blood cholesterol levels and APOE‐ɛ4 status on log‐transformed WMH volume. Associations between cholesterol markers and WMH volume also were separately examined in APOE‐ɛ4 carriers and non‐carriers. Covariates included age, sex, years of education, intracranial volume, MRI scanner, body mass index, diabetes, and hypertension. All continuous variables were standardized. We corrected for three comparisons (HDL‐c, LDL‐c, triglycerides) using the false discovery rate method.

There was a significant HDL‐c×APOE‐ɛ4 interaction on WMH volume (β= ‐0.10, p‐corrected= 0.03) in the fully‐corrected model driven mainly by a non‐significant association between higher HDL‐c and lower WMH volume in APOE‐ɛ4 carriers (β= ‐0.10, p‐corrected= 0.17) only. Interactions between APOE4 carrier status and LDL‐c and triglycerides were not significant.

APOE‐ɛ4 moderated the relationship between HDL‐c and WMH volume such that greater HDL‐c levels were associated with lower WMH burden in ɛ4 carriers only. APOE‐ɛ4 did not interact with LDL‐c or triglycerides on WMH volume. This investigation provides support for investigating HDL‐c further in the context of brain health in ɛ4 carriers.

Figures