In APOEε4 carriers, estradiol use is associated with a healthier brain than no hormone therapy, while non-carriers show no such association: a large study explains why evidence on menopausal hormone therapy and the brain has been contradictory for decades
In APOEε4 carriers, estradiol use is associated with a healthier brain than no hormone therapy, while non-carriers show no such association: a large study explains why evidence on menopausal hormone therapy and the brain has been contradictory for decades
A study of 8741 women from the NACC database, a dementia data repository spanning 40+ centers across the United States, found that the association between menopausal hormone therapy and memory and brain structure depends on both the drug formulation and APOEε4 carrier status. Among carriers of this gene, estradiol was associated with more favorable outcomes than the classic conjugated estrogen formulation; among non-carriers, the difference was barely detectable.
In 2003, the memory arm of the Women's Health Initiative, a large U.S. women's health program, found that women aged 65 and older taking conjugated equine estrogens with medroxyprogesterone (the drugs Premarin and Prempro) had a nearly doubled risk of probable dementia, with a hazard ratio of 2.05. Later observational studies using estradiol pointed in the opposite direction, showing a memory benefit, and some hinted at a role for the APOEε4 gene: among carriers, therapy was sometimes associated with a lower dementia risk. Without accounting for formulation, this clue could not explain the conflicting results, and the debate continued for twenty years.
In a preprint published on September 20, the laboratory of Liisa Galea at the CAMH psychiatric center in Toronto compares formulations head to head in a single cohort. Participants (mean age approximately 70 years) were divided into five groups: no therapy, estradiol (the most potent endogenous human estrogen) alone or with a progestogen, and conjugated equine estrogens (CEE, an extract from pregnant mare urine, the same Premarin and Prempro) alone or with medroxyprogesterone.
Those taking CEE had worse episodic memory than those on estradiol or those receiving no therapy, but CEE was associated with better verbal fluency. After statistical balancing of the groups by age and risk factors, a naming test revealed an interaction between formulation and genotype: estradiol conferred a greater benefit in APOEε4 carriers, while conjugated estrogens conferred a greater benefit in non-carriers.
The difference lies in pharmacology: CEE consists primarily of a weak estrogen with low affinity for brain receptors, whereas estradiol binds them far more strongly and has long been shown to protect the hippocampus, the structure responsible for episodic memory. The liver converts ingested estradiol to a weaker form; a patch or gel bypasses the liver.
The benefit of estradiol was more pronounced among Black participants than among White participants, even though in the United States Black women experience earlier and more severe menopause but receive hormone therapy less often. Among 165 women with data from before and after discontinuing therapy, APOEε4 carriers who stopped estradiol lost their advantage on the naming test; after discontinuing CEE, episodic memory declined across all groups.
The strongest confirmation came from MRI: brain scans were available for 930 participants, 871 without therapy and 59 on estradiol. In APOEε4 carriers taking estradiol, the cortex was thicker in four regions that are among the first affected in Alzheimer's disease, the hippocampus was larger, and white matter hyperintensities (a marker of cerebrovascular damage) were fewer than in carriers not receiving therapy. Non-carriers showed no such association.
The authors frame their conclusion as follows:
These results reframe the contradictory evidence on menopausal hormone therapy as a question of what should count as the same treatment, and support an approach to menopausal care that accounts for both drug formulation and genetic risk.