Investigating the impact of sex and reproductive aging on latent signatures of modifiable dementia risk factors.

Treating modifiable risk factors for dementia may serve to prevent maladaptive brain aging. These include, and are not limited to, obesity, hypertension, and other metabolic disorders. Nearly two-thirds of affected individuals are female, and emerging research has pointed to potential sex-specific factors, such as reproductive aging, as potent modifiers of dementia risk. Here, we leverage neuroimaging to characterize sex differences in the neural signatures of modifiable dementia risk factors, examining their relationship with brain anatomy, and integrate female-specific factors related to menopause history in a sex-specific analysis. Using cross-sectional data from the UK Biobank, we selected a cohort of 31,711 (age 44-82, 55.2% female/44.8% male) participants without diagnosed neurological conditions and collated behavioral data previously determined as modifiable risk factors. Using partial least squares analysis (PLS), we examined latent signatures that represent linear combinations that maximize covariance between patterns of brain (mean cortical thickness from 64 regions) and risk factor variables. To examine sex differences, we performed PLS analysis using the entire sample. We performed linear models to explore age-by-sex interactions with the PLS-derived brain scores and the risk factor pattern scores. To examine sex-specific relationships, we performed separate PLS analyses for the males and females and integrated menopause-related variables into the latter analysis. Our study found sex-dependent and menopause-dependent relationships between lifestyle risk factors and cortical thickness, highlighting stronger impacts of cardiometabolic factors on males and social and obesity-related factors on females in preserving brain health. The inclusion of menopause-related variables did not change the relationships with lifestyle risk factors, and strict age matching dampened the strength of the findings. Our findings suggest that lifestyles and the female-specific endocrine environment influence sex differences in cortical anatomy during brain aging.