AUTHOR=Nerattini Matilde , Rubino Federica , Jett Steven , Andy Caroline , Boneu Camila , Zarate Camila , Carlton Caroline , Loeb-Zeitlin Susan , Havryliuk Yelena , Pahlajani Silky , Williams Schantel , Berti Valentina , Christos Paul , Fink Matthew , Dyke Jonathan P. , Brinton Roberta Diaz , Mosconi Lisa TITLE=Elevated gonadotropin levels are associated with increased biomarker risk of Alzheimer's disease in midlife women JOURNAL=Frontiers in Dementia VOLUME=2 YEAR=2023 URL=https://www.frontiersin.org/journals/dementia/articles/10.3389/frdem.2023.1303256 DOI=10.3389/frdem.2023.1303256 ISSN=2813-3919 ABSTRACT=Introduction

In preclinical studies, menopausal elevations in pituitary gonadotropins, follicle-stimulating hormone (FSH) and luteinizing hormone (LH), trigger Alzheimer's disease (AD) pathology and synaptic loss in female animals. Herein, we took a translational approach to test whether gonadotropin elevations are linked to AD pathophysiology in women.

Methods

We examined 191 women ages 40–65 years, carrying risk factors for late-onset AD, including 45 premenopausal, 67 perimenopausal, and 79 postmenopausal participants with clinical, laboratory, cognitive exams, and volumetric MRI scans. Half of the cohort completed 11C-Pittsburgh Compound B (PiB) amyloid-β (Aβ) PET scans. Associations between serum FSH, LH and biomarkers were examined using voxel-based analysis, overall and stratified by menopause status. Associations with region-of-interest (ROI) hippocampal volume, plasma estradiol levels, APOE-4 status, and cognition were assessed in sensitivity analyses.

Results

FSH levels were positively associated with Aβ load in frontal cortex (multivariable adjusted P ≤ 0.05, corrected for family wise type error, FWE), an effect that was driven by the postmenopausal group (multivariable adjusted PFWE ≤ 0.044). LH levels were also associated with Aβ load in frontal cortex, which did not survive multivariable adjustment. FSH and LH were negatively associated with gray matter volume (GMV) in frontal cortex, overall and in each menopausal group (multivariable adjusted PFWE ≤ 0.040), and FSH was marginally associated with ROI hippocampal volume (multivariable adjusted P = 0.058). Associations were independent of age, clinical confounders, menopause type, hormone therapy status, history of depression, APOE-4 status, and regional effects of estradiol. There were no significant associations with cognitive scores.

Discussion

Increasing serum gonadotropin levels, especially FSH, are associated with higher Aβ load and lower GMV in some AD-vulnerable regions of midlife women at risk for AD. These findings are consistent with preclinical work and provide exploratory hormonal targets for precision medicine strategies for AD risk reduction.