Abstract
BACKGROUND: Cerebrovascular disease (CVD) is a major cause of mortality in females, while two‐thirds of Alzheimer’s disease (AD) patients are female. AD and CVD share many genetic risk factors, one of them being apolipoprotein E (APOE) genotype. Sex differences in APOE and AD are well‐established; it is unclear if associations between APOE and CVD are sex‐specific. METHOD: Autopsy data (females = 3,464; males = 3,130) from three AD cohorts included measures of arteriolosclerosis and atherosclerosis (levels for each: none [0], mild [1], moderate [2], and severe [3]) harmonized as part of the AD Sequencing Project Phenotype Harmonization Consortium. Non‐Hispanic White (NHW) individuals >60 years at age of death were included (average age at death = 85±9; 58% AD dementia). Linear analyses assessed sex‐stratified and APOE*sex interactions on CVD outcomes covarying for age at death. APOE‐ε4 and APOE‐ε2 were modeled additively and dominantly, respectively. RESULT: Females were older on average and included more APOE‐ε2 and less APOE‐ε4 carriers when compared to males. Among males and females, APOE‐ε2 was not related to arteriolosclerosis (p = 0.90) or atherosclerosis (p = 0.30), while APOE‐ε4 related to higher arteriolosclerosis (β = 0.07 p<0.001) and atherosclerosis (β = 0.04, p = 0.03). There was a significant APOE‐ε2*sex interaction on arteriolosclerosis (p = 0.02) whereby the ε2 allele was associated with less arteriolosclerosis among females (β = ‐0.05, p = 0.27) and more arteriolosclerosis among males (β = 0.08, p = 0.13), although neither stratified effect reached statistical significance. Similarly, the ε4 allele was associated with more arteriolosclerosis among females (β = 0.11, p<0.001), but not among males (β = 0.04, p = 0.16), although the APOE‐ε4*sex interaction did not reach statistical significance (interaction‐p = 0.052). The ε4 allele was associated with more atherosclerosis among females (β = 0.07, p = 0.01), but not males (β = 0.01, p = 0.56), but again the APOE‐ε4*sex interaction did not reach statistical significance (p = 0.28). CONCLUSION: We provide evidence that the ε4 allele is related to higher levels of arteriolosclerosis and atherosclerosis, while providing modest evidence of a sex difference in the association between the ε2 allele and arteriolosclerosis. Future work will seek to better characterize these sex‐specific effects within the context of other neurodegenerative pathologies in the aging brain, and more fully characterize sex‐specific clinical consequences of APOE and CVD.