Women who used estrogen-only menopausal hormone therapy (MHT) had 35% lower odds of Alzheimer’s disease (AD) neuropathologic change at autopsy and 39% lower odds of clinician-defined dementia in a large observational study led by Stanford researchers.

The study analyzed data from two cohorts: the National Alzheimer’s Coordinating Center (NACC) autopsy cohort and the Alzheimer’s Disease Neuroimaging Initiative (ADNI) living cohort. Researchers also found associations between MHT use and higher amyloid-beta levels in blood and cerebrospinal fluid (CSF), biomarkers associated with less amyloid deposition in the brain.

The findings, published in Neurology, add to decades of conflicting evidence on hormone therapy and cognitive health.

“In light of so many studies that have focused on more subjective measures like cognitive decline, we focused on the actual brain tissue, the gold standard for measuring pathological brain changes, and combined that with cognition and biomarkers measured while the women were alive,” said lead study author Jennifer Bruno, PhD, instructor in psychiatry and behavioral sciences at Stanford University School of Medicine in Stanford, California.

“However, the study is observational, so we cannot conclude that the relationship is causal or that estrogen therapy prevents Alzheimer’s disease,” she told Medscape Medical News.

AD Pathology

Both cohorts included women aged 50 years or older; those using combined estrogen-progestin therapy were excluded. In the NACC autopsy cohort, 258 women reported estrogen-only MHT use, and 2701 did not. Estrogen-only MHT use was linked to lower odds of greater AD neuropathologic change, measured by the National Institute on Aging-Alzheimer’s Association ABC score (odds ratio [OR], 0.65; 95% CI, 0.48-0.88). The ABC score captures three hallmarks of AD: amyloid plaques, neurofibrillary tangles, and neuritic plaque density.

The association persisted after adjusting for age, education, race, hypertension, and APOE epsilon 4 status, a major genetic AD risk factor. However, the association was statistically significant only in women without the APOE epsilon 4.

A secondary subgroup analysis showed a stronger association with conjugated estrogen (OR, 0.55; P = .013) than with estradiol, which was not statistically significant (OR, 0.72; P = .086). “The difference does not mean that one formulation is better than the other,” said coauthor Hadi Hosseini, PhD, associate professor of psychiatry and behavioral sciences at Stanford.

“For both types, we saw the association in the same direction,” he told Medscape Medical News. “The lack of statistical significance for estradiol may be due to smaller sample size.”

Biomarkers and Clinical Outcomes

In the living ADNI cohort, comprising 110 MHT users and 1948 nonusers, estrogen-only therapy was associated with higher plasma amyloid-beta 42/40 levels (beta = 0.44; 95% CI, 0.16-0.73; P = .0025) and CSF amyloid-beta 1-42 levels (beta = 0.07; 95% CI, 0.002-0.13; P = .030). “There are several mechanisms that could explain a protective effect of estrogen in the brain,” said Hosseini. “Estrogen may reduce amyloid production, enhance its clearance, inhibit tau-related phosphorylation, and exert anti-neuroinflammatory effects.”

Clinical outcomes were also favorable, though less consistent across cohorts. In NACC, MHT use was linked to lower odds of clinician-defined dementia (OR, 0.61; 95% CI, 0.55-0.67) and memory or functional decline (OR, 0.67; 95% CI, 0.61-0.74; both P < .0001). In ADNI, it was linked instead to better verbal memory, including immediate recall (beta = 0.336; P = .002) and learning (beta = 0.233; P = .033).

Study Limitations and Caution

The study was limited to estrogen-only therapy, which is generally used in women who have had a hysterectomy, and did not examine combined estrogen-progestin therapy. The findings may therefore not apply to women using combined MHT.

The researchers could not assess whether treatment timing or duration influenced the findings because they lacked data on when women started MHT, how long they used it, or whether they switched formulations. Previous research suggests these factors may affect cognitive outcomes.

A subgroup of women who started therapy before age 60 showed a similar pattern, Bruno said, although the analysis remained observational. Bruno and Hosseini said prospective studies following women from the menopausal transition are needed to determine whether the timing or duration of MHT influences AD-related outcomes.

Pauline Maki, PhD, professor of psychiatry, psychology, and obstetrics and gynecology at the University of Illinois in Chicago, who was not involved in the study, took a more critical view. “This study by design is highly limited in its ability to even begin to address the question,” she told Medscape Medical News.

“The randomized trial data to date conflict with the conclusions in this study.” Maki referenced a recent Women’s Health Initiative analysis, which found no differences in AD biomarkers among women randomized to hormone therapy or placebo.

She also raised concerns about “healthy-user bias,” noting that MHT users may differ from nonusers in healthcare access, education, and cardiovascular health. She pointed to population-based studies from countries with more equitable healthcare access that have likewise found no protective effect.

“In databases of entire countries, such as those in Scandinavia or Taiwan, we see no benefit for reduced risk of Alzheimer’s disease,” she said.

The study was supported by the National Institute on Aging. The authors reported no relevant financial disclosures.