We have been telling the wrong story about Alzheimer’s and hormones.
For years the conversation went like this: estrogen declines during menopause, the brain loses protection, cognitive function deteriorates. Replace estrogen and maybe you slow the damage. That framing shaped research, clinical guidelines, and how most of us were trained to think about cognitive aging in women. It felt logical. It was also incomplete.
A 2025 study published in Nature says the variable we should have been watching is not estrogen decline. It is the ratio of estradiol to progesterone during perimenopause. And the mechanism it maps is not vague hand-waving about neuroprotection. It is specific: receptor destabilization, mitochondrial energy failure, glutamate excitotoxicity, and neurodegeneration. A cascade. One that begins years before a woman ever reaches menopause.
I break down the full mechanism, the clinical implications, and what these changes mean for prescribing in Episode 94 of the HRT University Podcast.
| Quick Answer A 2025 Nature study found that Alzheimer’s risk in women is not driven by estrogen decline alone. The imbalance between estradiol and progesterone during perimenopause destabilizes estrogen receptor alpha, disrupts mitochondrial energy metabolism via ERR alpha, and triggers excitotoxic neuronal damage. Restoring bioidentical progesterone, not synthetic progestins, is essential for stabilizing this pathway and supporting long-term neuroprotection. |
Why Are Two-Thirds of Alzheimer’s Patients Women?
This is the question that should have reframed the entire field a long time ago. Roughly two-thirds of all Alzheimer’s patients are women. The old explanation was longevity: women live longer, so they accumulate more neurodegeneration. That was a statistical observation dressed up as a cause. It never explained the mechanism.
Here is what actually happens during perimenopause. Estradiol does not decline in a clean, predictable line. It fluctuates wildly. Some months it spikes above premenopausal levels. Other months it craters. Meanwhile, progesterone begins a steady, sustained drop. The result is a prolonged period of relative estrogen excess, where the ratio of estradiol to progesterone tips heavily toward estradiol.
This is the setup that everything else follows from. It is not a decline. It is an imbalance. And the brain responds to the imbalance before either hormone fully disappears.
If you have sat with a perimenopausal patient whose labs look “normal” but whose cognition, mood, and sleep tell a completely different story, this is part of what you are seeing. The numbers on the page do not capture what the ratio is doing inside the brain.
How the Estradiol-to-Progesterone Imbalance Damages the Brain
The Nature study maps a specific metabolic cascade. Each step follows from the one before it.
Estrogen receptor alpha destabilization.
Without sufficient progesterone to counterbalance circulating estradiol, ERα in the brain becomes structurally unstable. This is not a minor receptor issue. ERα regulates cholesterol synthesis, membrane integrity, and synaptic function. When it destabilizes, the downstream effects are not subtle.
ERR alpha dysfunction.
ERα instability disrupts a downstream protein called ERR alpha. This is not a hormone receptor. It is a protein that regulates mitochondrial energy metabolism, specifically glucose utilization in neurons. If you have been through the Intro Module of the Master Course, this connects directly to the metabolic foundation: when mitochondrial function is compromised, the cell shifts from oxidative phosphorylation toward glycolysis. Less ATP. More waste. Less resilience.
Mitochondrial energy failure.
When ERR alpha is compromised, neurons lose their capacity to process glucose efficiently. The brain is the most energy-demanding organ in the body. It does not tolerate energy deficits well. And unlike a muscle that can switch fuel sources, the brain is exquisitely dependent on glucose metabolism. Localized energy failure follows.
Glutamate excitotoxicity.
Energy-starved neurons increase glutamate activity as a compensatory response. Think of it this way: one cup of coffee is productive. Five cups is destructive. Glutamate at normal levels is essential for signaling. In excess, it becomes neurotoxic. That excitatory overload, combined with energy failure, initiates neuronal damage.
Neurodegeneration begins.
This is not an overnight event. It accumulates over years during perimenopause while the hormonal imbalance persists. By the time a woman reaches full menopause, the critical damage window may have already passed.
Every step in this cascade traces back to the same thing the Master Course teaches about metabolic dysfunction: when energy production fails, structure degrades. Disease follows. The bridge analogy holds here. Remove the energy, and the structure collapses.
Why Estrogen Alone Is Not Neuroprotective
This is the reframe that matters most clinically.
Unopposed estradiol, given without adequate progesterone, can actually worsen the neuronal environment by perpetuating the ratio imbalance that drives ERα instability. Estrogen is not harmful. But estrogen without progesterone is incomplete. In the context of Alzheimer’s prevention, it may be counterproductive.
I will say this directly: the estrogen-is-queen framing needs to stop dominating the conversation for women. Estradiol is critical. Nobody is dismissing it. But if you are prescribing estradiol without progesterone and calling it neuroprotection, you may be doing more harm than good. The data now supports that position. This is not opinion. It is mechanism.
Think about this through the lens of estrogen as a class, which the Master Course covers in the Big Five. Contained in the fireplace, estrogen warms. In the living room, it destroys. Without progesterone to contain it, the balance tips toward proliferation, inflammation, and in this specific context, receptor destabilization in the brain.
What Bioidentical Progesterone Does in the Brain
Progesterone is not merely a reproductive hormone. It is a potent neurosteroid with direct effects on brain tissue. If you have only been prescribing it for endometrial protection, you are working with half the picture.
Progesterone stabilizes ERα, preventing the cascade described above. It supports cholesterol synthesis in the brain, which is essential for healthy cell membranes, synapse formation, and local steroid hormone production. It preserves mitochondrial function by maintaining ERR alpha activity and neuronal glucose metabolism.
And then there is allopregnanolone.
Progesterone converts to allopregnanolone via 5-alpha reductase. Allopregnanolone is a positive allosteric modulator of GABA-A receptors, binding at a separate site from benzodiazepines. It reduces excitotoxicity, supports neuronal repair, and modulates the stress response. This metabolite is one of the key mechanisms through which progesterone delivers neuroprotection. It is also why the FDA approved brexanolone in 2019 and zuranolone in 2023, both of which mimic allopregnanolone’s GABA activity. The system approved drugs that replicate what bioidentical progesterone already does, while still prescribing synthetic progestins that block this pathway entirely.
The Female Module, Section 2 of the Master Course covers progesterone’s neuroprotective, mitochondrial, and GABAergic effects. The Advanced Female Module, Section 4 covers allopregnanolone pharmacology in detail.
Why Synthetic Progestins Cannot Replace Bioidentical Progesterone
Synthetic progestins such as medroxyprogesterone acetate do not share the same metabolic pathways as bioidentical progesterone. MPA is derived from testosterone. It is not progesterone. It is not even a hormone. It binds the progesterone receptor uncleanly, and it also hits androgen, glucocorticoid, and mineralocorticoid receptors. It does not convert to allopregnanolone. Some progestins actively oppose the neuroprotective mechanisms that endogenous progesterone supports.
When a provider prescribes MPA for endometrial protection during perimenopause, the uterus may be addressed, but the brain is not receiving what it needs. It may function for her endometrial lining, but it is not going to do a whole lot for her brain. And we just talked about how more women than men end up with Alzheimer’s because of this very fact.
The December 2025 review in Frontiers in Global Women’s Health recommended removing MPA from the market entirely. Micronized progesterone has an odds ratio of 0.99 for breast cancer. Synthetic progestins come in at 1.28. The structure is different. The function is different. The clinical outcome is different. Structure equals function. Brick house versus straw house.
What This Means for Clinical HRT Prescribing
Start earlier.
The hormonal imbalance begins in perimenopause, often years before menopause is confirmed. Waiting means the neuronal damage has been accumulating. I estimate a 10- to 15-year window that is lost when providers take a wait-and-see approach. By the time someone presents with cognitive symptoms, the critical intervention period may have already closed.
Progesterone is not an afterthought.
It should be part of the hormonal strategy from the beginning, dosed for neuroprotection, not just endometrial safety. If your reasoning for adding progesterone is “she still has a uterus,” you are missing the larger clinical picture.
The ratio matters.
High-dose estradiol without proportional progesterone does not constitute adequate neuroprotection. This is not about estrogen being bad. It is about the relationship between two hormones that the brain depends on working together.
Route and formulation matter.
Oral, sublingual, and vaginal bioidentical progesterone each have different absorption profiles. Compounding allows individualized dosing tailored to the patient. This ties directly to the compounding pharmacy argument from Episode 92.
HRT is preventive medicine.
This study provides a clear biological rationale for framing hormone replacement therapy as foundational, not optional. Not symptom management. Not a convenience. A metabolic intervention that changes the trajectory of neurodegeneration before it becomes irreversible.
Key Takeaways
- Alzheimer’s risk in women is driven by the estradiol-to-progesterone ratio during perimenopause, not estrogen decline alone.
- The cascade runs from ERα instability to ERR alpha dysfunction to mitochondrial failure to glutamate excitotoxicity to neurodegeneration.
- Unopposed estradiol can worsen the problem. Progesterone is required for the full neuroprotective effect.
- Synthetic progestins do not convert to allopregnanolone and cannot replicate bioidentical progesterone’s brain effects.
- Perimenopause is the critical intervention window. Waiting until postmenopause or cognitive symptoms may be 10 to 15 years too late.
- HRT is not symptom management. It is foundational preventive medicine with clear biological rationale.
FAQ
Does estrogen cause Alzheimer’s disease?
No. Estrogen itself does not cause Alzheimer’s. The 2025 Nature study shows that the imbalance between estradiol and progesterone is the metabolic trigger. Estrogen without adequate progesterone can worsen the neuronal environment, but that is a problem of imbalance, not estrogen toxicity.
What is the estradiol-to-progesterone ratio and why does it matter?
During perimenopause, estradiol fluctuates erratically while progesterone declines steadily. This creates sustained relative estrogen excess. The Nature study identifies this ratio imbalance as the driver of ERα instability, which triggers downstream mitochondrial dysfunction and neuronal damage.
Is progesterone neuroprotective?
Yes. Progesterone stabilizes ERα, supports brain cholesterol synthesis, preserves mitochondrial energy metabolism, and converts to allopregnanolone, a neurosteroid that modulates GABA and promotes neuronal repair. These effects are specific to bioidentical progesterone, not synthetic progestins.
Can synthetic progestins like MPA protect the brain?
No. Medroxyprogesterone acetate does not convert to allopregnanolone and does not share the neuroprotective metabolic pathways of bioidentical progesterone. MPA binds the progesterone receptor uncleanly and also activates androgen, glucocorticoid, and mineralocorticoid receptors. Some progestins may actively oppose protective mechanisms.
When should bioidentical progesterone be started for Alzheimer’s prevention?
The hormonal imbalance begins during perimenopause, often years before menopause is confirmed. Starting bioidentical progesterone early addresses the ratio imbalance before significant neuronal damage accumulates. Waiting until postmenopause or cognitive symptoms may mean the critical window has already closed.
What is ERR alpha?
ERR alpha (estrogen-related receptor alpha) is a protein that regulates mitochondrial energy metabolism, specifically glucose utilization in neurons. When ERα becomes unstable due to hormonal imbalance, ERR alpha function is compromised, leading to energy failure in brain cells. This connects directly to the metabolic foundation taught in the Master Course: when energy production fails, cellular structure degrades.
What is allopregnanolone?
Allopregnanolone is a metabolite of progesterone produced via 5-alpha reductase. It is a positive allosteric modulator of GABA-A receptors, binding at a separate site from benzodiazepines. It reduces excitatory stress, supports neuronal repair, and modulates the stress response. Synthetic progestins do not produce this metabolite. The FDA approved brexanolone (2019) and zuranolone (2023) to mimic its effects, while the system still prescribes progestins that block this pathway.
Does this mean estrogen is unnecessary for brain health?
No. Estrogen remains important. The study’s contribution is showing that estrogen alone is insufficient. The balance between estradiol and progesterone is what stabilizes the receptor and metabolic pathways that protect against Alzheimer’s. Both hormones are necessary. Neither one is the whole answer.
Listen to Episode 88 about Progesterone and Brain Health.
Why perimenopause and not menopause?
Perimenopause is when the ratio is most disrupted. Estradiol is erratic and sometimes elevated. Progesterone is steadily declining. This creates the sustained imbalance that drives ERα destabilization. By full menopause, both hormones have declined, but the critical damage window, the period of maximal ratio disruption, may have already passed. That is why early intervention matters.
Where was this study published?
Nature, 2025. Full title: “Perimenopausal State Estradiol-to-Progesterone Imbalance Drives Alzheimer’s Risk via ERR Alpha Dysregulation and Energy Dyshomeostasis.”
