A woman in her early sixties sat across from me last month, holding a list of questions she had typed out the night before. She had been on estradiol and progesterone for three years under my supervision, and her internist had just told her she should stop. "You're past the window," he said. "The risks outweigh the benefits." She wanted to know if that was true.
I told her what I have told patients for years: the conventional window-of-opportunity narrative is incomplete, and the brain data increasingly supports what we have been doing at The Neurogenesis Project. Three studies published this year make that case more clearly than anything before them.
The Vascular Evidence from Gothenburg
On November 4, a team led by Anna Marseglia at the Karolinska Institutet published results from 352 cognitively healthy women in Gothenburg, all aged 70, whose brains were scanned with MRI at enrollment and again five years later. The study, published in Neurology, measured the relationship between lifetime estrogen exposure and cerebral small vessel disease, the slow accumulation of damage in the brain's tiniest blood vessels that contributes to cognitive decline, vascular dementia, and stroke risk.
The findings were clear. A longer reproductive period, measured as the years between first menstruation and menopause, was associated with 8 percent lower odds of having two or more markers of cerebral small vessel disease. Women who had used menopausal hormone therapy showed less small vessel disease and more favorable white matter microstructure on diffusion imaging. And the women who had used both oral contraceptives during their reproductive years and hormone replacement therapy during menopause showed the most favorable vascular brain health profile of the entire cohort.
Marseglia's group was careful to note that the study is observational and cannot establish cause and effect. That is the correct scientific posture. But the direction of the data is consistent, and it aligns with something I wrote about in May: menopause produces measurable neurodegeneration in frontal and temporal brain regions, and estrogen loss is a primary driver.
What Stanford Found in Autopsied Brains
Three months earlier, Hadi Hosseini's team at Stanford Medicine published a study in Neurology that took a different approach entirely. Instead of imaging living brains, they examined the actual pathology in autopsied ones. Their dataset included 21,462 participants across two existing research databases, with 258 brain autopsies of women who had used estrogen-only menopausal hormone therapy compared against 2,701 women who had not used any form of hormone therapy.
The autopsy data showed a 35 percent reduction in the odds of Alzheimer's-associated pathological hallmarks, including amyloid plaques and neurofibrillary tangles, in the estrogen-only users. Living participants' data supported the finding: estrogen-only users had 39 percent lower odds of receiving a clinical dementia diagnosis during their lifetimes and performed better on memory tests.
Jennifer Bruno, one of the study's lead authors, put it plainly: "For a long time, the going recommendation was 'Don't use MHT for memory decline.' This study flies in the face of that recommendation."
A caveat matters here. The Stanford study had enough autopsy data only for women on estrogen-only formulations, most of whom had had hysterectomies. The number of autopsied women on combined estrogen-plus-progestin regimens was too small to reach a conclusion. That distinction is important and I will come back to it.
The Mechanism in the Mouse Brain
A third piece of the picture came from Northwestern in May 2026. Piehl and colleagues published a preclinical study in Aging Cell showing that loss of brain-derived estrogen in female mice was associated with disruption of the extracellular matrix in the hippocampus, the brain region most responsible for memory formation. The extracellular matrix is the structural scaffolding that holds neurons in place and facilitates their communication. When estrogen production drops, the genes that maintain this scaffolding, including Dcn and Ogn, become dysregulated.
The finding that stood out: older female mice were uniquely sensitive to this loss, while males were not. The researchers used aromatase-deficient mice, lacking the enzyme required to produce estrogen either body-wide or only in the brain, and found that the combination of advanced age and female sex produced the most severe matrix deterioration. This is preclinical work, so it does not prove a human benefit. But it offers a plausible biological mechanism for why estrogen loss hits the female brain harder than the male one, and why replacing it might matter.
What We Have Been Doing at NGP
I want to be direct about why these studies matter to our practice. At The Neurogenesis Project, we have encouraged hormone replacement therapy in women well past the conventional "window," including women in their sixties and seventies, because the clinical evidence for cognitive preservation has been there for years. The new data from Karolinska and Stanford validates what we have been doing.
But the way we prescribe matters as much as whether we prescribe. The Stanford study could only draw conclusions about estrogen-only therapy, which is standard for women without a uterus. For women who still have a uterus, estrogen alone raises the risk of endometrial cancer. That is why we balance estradiol with progesterone in every case where the uterus is intact. This is not a minor detail. It is the difference between a therapy that protects the brain and one that creates a new problem.
We also do something most prescribers do not. We test blood hormone levels and their metabolites regularly to determine the dosing that is optimal for brain health in each individual patient, not just the dose that controls hot flashes. Estrogen metabolism matters because the way the body breaks down estradiol determines whether its byproducts are protective or potentially harmful. In our Intensive Brain Health Program, hormone optimization is one data stream among several, including neuroimaging, pharmacogenomics, and neurotransmitter metabolite testing, that together determine treatment. A single serum estradiol level drawn once a year is not sufficient to guide therapy aimed at neuroprotection.
This is what precision brain medicine looks like in practice. It is what the Neuroeconomy demands: treating cognitive capacity as the asset it is, and protecting it with the same specificity we would apply to any other organ system.
What This Means Going Forward
The convergence of these three studies, a population-based vascular imaging study, an autopsy-based pathology study, and a mechanistic animal study, all pointing in the same direction, is the kind of evidence pattern that moves the field. None of them alone is definitive. All of them are observational or preclinical. The randomized trial that everyone wants, testing whether initiating hormone therapy in a specific population at a specific time prevents dementia, has not been done and may take a decade to complete.
In my practice, I do not wait for that trial to treat the patients sitting in front of me. The evidence already supports a careful, monitored, individually dosed hormone replacement strategy aimed at preserving brain vascular health and reducing Alzheimer's-associated pathology. The new data makes that case stronger, not weaker.
If you are a woman who has been told that hormone therapy is too risky or that the window has closed, it may be worth revisiting that conversation with a physician who tracks the current literature and who is willing to monitor the therapy properly. That is what we do here.
This article was drafted with the assistance of AI writing tools, then reviewed, edited, and approved by Dr. Sean C. Orr, M.D., who holds full editorial responsibility for its content.
Sources:
Lorenzon, G., et al. "Reproductive lifespan and hormonal therapy in relation to later-life neurovascular health." Neurology, November 4, 2026. DOI: 10.1212/WNL.0000000000218706.
Bruno, J., Shaw, J., Hosseini, H., et al. "Estrogen-based menopausal hormone therapy and Alzheimer's risk." Neurology, August 12, 2026. DOI: 10.1212/WNL.0000000000218413.
Piehl, N.C., et al. "Loss of Brain-Derived Estrogen Is Associated With Sex- and Age-Dependent Alterations in Memory, Affective Behavior, and Hippocampal Extracellular Matrix Gene Expression." Aging Cell, May 2026.