Understanding Hormonal Balance and Brain Health in Perimenopause: New Mechanistic Insights

Everything in menopause research feels like a paradox: some studies say hormone therapy helps, others suggest it doesn’t, and clinicians keep asking “what’s the ideal level?” But a new Nature Communications paper (https://rdcu.be/e1l4F) finally gives us a mechanistic insight—and it points squarely at balance, not absolute hormone numbers.

Perimenopause Isn’t Just “Low Hormones”—It’s Imbalanced Signaling

The study modeled a perimenopausal state in mice where estradiol (E2) was elevated and progesterone (P4) was reduced, producing a high E2:P4 ratio. In humans, this mimics the erratic hormonal fluctuations of perimenopause—a time when estrogen spikes become more unpredictable and luteal progesterone falls off hard.

What the researchers found was striking: that hormonal imbalance disrupted ERRα signaling, a key regulator of neuronal energy metabolism and cholesterol homeostasis in the brain. This leads to more glutamate release, increased neuronal excitability, ATP depletion and lower energy resilience, and greater vulnerability to stressors that drive neurodegeneration.

In short: it’s not estrogen by itself that’s risky—it’s estrogen without progesterone to properly buffer it.

Think in Ratios, Not Absolutes

In fields like lipid science, we talk about EPA:DHA ratios because balance determines function, not total amount. The same principle may apply to reproductive hormones. This doesn’t mean we have a numerical target (like “ideal E2:P4 = X”), but it does suggest the brain is sensitive to the proportion of excitatory (E2) versus inhibitory (P4) signaling.

Why Progesterone Matters—Especially in the Brain

Progesterone isn’t just about protecting the uterus. It modulates estrogen receptor activity, enhances inhibitory neurotransmission, supports mitochondrial energy pathways, and counters excitotoxic processes.

In the Nature Communications model, progesterone helped maintain ERRα activity, tying it to energy metabolism and preventing the metabolic “crisis” that increases Alzheimer’s vulnerability. Without it, estradiol’s effects become unmoderated—much like spinning the wheels of a car without traction.

However, progesterone’s role is more complex than simple “buffering.” The literature shows that bioidentical progesterone may support neuroprotection through BDNF upregulation and maintenance of ERRα function, while synthetic progestins like medroxyprogesterone acetate (MPA) can antagonize estrogen’s protective effects by down-regulating ERβ and BDNF. Preclinical studies demonstrate that cyclic progesterone exposure preserves or enhances estradiol’s neuroprotective effects, whereas continuous progesterone exposure can blunt or reverse estradiol-mediated benefits on amyloid burden, synaptic plasticity, and neuronal survival.

This distinction is critical: observational studies and RCTs show that estrogen-progestogen therapy (typically using synthetic progestins) is associated with increased dementia risk, while estrogen-only therapy shows protective effects when initiated in midlife. The Nature Communications findings about progesterone maintaining ERRα function likely apply to bioidentical progesterone, not necessarily to all progestogen formulations used clinically.

No clinical trials have directly compared bioidentical estrogen-progesterone therapy versus estrogen combined with progesterone-only IUD formulations in APOE4-positive perimenopausal women. This represents a critical evidence gap given the mechanistic insights from the Nature Communications paper about E2:P4 balance and ERRα signaling. Furthermore, Levonorgestrel does reach the brain and has CNS effects, but these effects are qualitatively different from—and in some cases opposite to—bioidentical progesterone's neuroprotective actions. The concern is not that levonorgestrel from an IUD doesn't reach the brain, but that its effects on energy metabolism, neurogenesis, and allopregnanolone synthesis may not provide the same neuroprotective "buffering" that bioidentical progesterone offers. And yet, the current medical guidance tells women “An IUD is adequate protection to commence estrogen replacement therapy unopposed by any further progesterone supplementation.”

What This Study Doesn’t Tell Us (Yet)

It does not give us a validated ratio or lab target to aim for. There is currently no clinical evidence that specific serum E2:P4 ratios reduce Alzheimer’s risk in humans.

What this paper does provide is a plausible mechanistic explanation for why:

1. Women have higher Alzheimer’s incidence

2. Risk spikes around perimenopause

3. Hormone therapy effects have been inconsistently positive or negative in past trials

It suggests that context—timing + balance + formulation + metabolic support—matters hugely.

How This Helps Clinicians and Women Think Differently

Instead of asking: “How much estrogen should a woman have?”

We should be asking:

• “Is estrogen being counterbalanced with sufficient progesterone—especially in the brain?”

• “Is the hormonal milieu supporting energy homeostasis, not excitatory overload?”

• “Are we using bioidentical hormones or synthetic formulations?”

This is why the study focused not on absolute concentrations but on system behavior—the neural metabolic networks that get dysregulated when the ratio skews.

A New Framework for Hormone Balance

Here’s a provisional way to think about this:

• 🌿 Balanced Zone – Estradiol and progesterone move in harmony, supporting ERRα and stable energy metabolism

• ⚖️ Ratio Drift Zone – Estradiol is high relative to progesterone. Neural energy systems are taxed. Risk for excitotoxic stress increases

• 🔥 Imbalance Zone – Long periods of disproportionate estradiol signaling with insufficient progesterone support may impair energy homeostasis and amplify vulnerability

This isn’t a specific lab threshold—it’s a functional concept built from mechanistic evidence. With time, and the right research, I do hope that we do develop specific lab thresholds…and as we will see below, they may differ depending on the woman’s Apoe4 gene.

Special Considerations for APOE4-Positive Perimenopausal Women

How to Interpret the Nature Communications Study in the Context of APOE4

The Nature Communications paper demonstrates that hormonal imbalance—specifically a high estradiol-to-progesterone ratio—disrupts ERRα signaling, impairs neuronal energy metabolism, and increases vulnerability to neurodegeneration. For APOE4-positive perimenopausal women, this mechanistic insight takes on heightened clinical significance.

APOE4 carriers experience accelerated endocrine and bioenergetic brain aging during midlife, with failure to mount adaptive bioenergetic reprogramming, significant mitochondrial dysfunction, increased neuroinflammation, and demyelination.APOE4-positive women also experience earlier menopause, and those with both APOE4 and early menopause exhibit the highest Alzheimer’s risk.

The interaction between APOE4 and menopause appears to operate through multiple converging pathways. Menopausal estrogen loss amplifies negative APOE4 effects, resulting in greater cognitive impairment in women compared to men, particularly affecting memory and language domains. Additionally, elevated follicle-stimulating hormone (FSH) in postmenopausal APOE4 females—but not APOE3 carriers—drives AD-like pathology through C/EBPβ/δ-secretase signaling, stimulating amyloid-β production and neurofibrillary tangles.

This creates a “triple hit” scenario: compromised bioenergetics from menopause, chronic low-grade inflammation from aging, and APOE4-related dyslipidemia and immune dysregulation. The combination of ovarian failure and APOE4 genotype exacerbates cognitive deficits and synaptic plasticity impairment beyond either factor alone.

Evidence-Based Advice for APOE4-Positive Perimenopausal Women

Midlife represents a critical intervention window for APOE4-positive women, as this is when accelerated brain aging processes begin and when preventive strategies may be most effective.

Hormone Therapy Considerations

The evidence for hormone therapy in APOE4 carriers reveals profound contradictions that underscore the critical importance of formulation, timing, and route of administration.

Supportive findings: Multiple studies show that APOE4 carriers using hormone therapy demonstrate improved delayed memory scores, 6-10% larger entorhinal and amygdala volumes, and longer leukocyte telomere length (suggesting reduced cellular aging) compared to non-users. Earlier initiation of hormone therapy was associated with larger hippocampal volumes only in APOE4 carriers, not in non-carriers. A pilot RCT found that transdermal 17β-estradiol in recently postmenopausal women was associated with reduced amyloid-β deposition, particularly in APOE4 carriers.

Concerning findings: Some observational studies show hormone therapy lowered cognitive impairment risk by nearly 50% in non-carriers but showed no effect in APOE4 carriers. Other studies suggest current hormone therapy users who are APOE4 carriers showed faster rates of cognitive decline. Recent data indicate that high postmenopausal estradiol concentrations were associated with older brain age in APOE4 carriers but younger brain age in non-carriers, suggesting APOE4 carriers may be particularly sensitive to excessive estrogen exposure without adequate progesterone buffering. However, this UK Biobank brain aging analysis study did not measure progesterone levels, did not measure hormone therapy, and did not directly test brain tissue physiology — only imaging-based brain age and circulating estradiol levels.

Formulation is critical: The contradictory findings likely reflect differences in hormone formulations. Bioidentical progesterone increases BDNF expression and provides neuroprotection, while medroxyprogesterone acetate (MPA) does not. Studies showing harm often used conjugated equine estrogens (CEE) plus MPA, while those showing benefit often used bioidentical estradiol. Estrogen-progestogen therapy (typically using synthetic progestins) is associated with increased Alzheimer’s risk when used for 5-10+ years, while estrogen-only therapy shows protective effects.

Route of administration matters: Transdermal 17β-estradiol appears particularly beneficial in APOE4 carriers, showing reduced amyloid deposition. Transdermal and vaginal estradiol bypass hepatic metabolism, providing steady-state concentrations rather than fluctuating levels.

Practical hormone therapy recommendations for APOE4 carriers:

• Bioidentical formulations: Estradiol (not CEE) plus bioidentical progesterone (not MPA) if a uterus is present

• Transdermal or vaginal estradiol: To provide steady-state levels and avoid hepatic first-pass metabolism

• Initiation during perimenopause or early postmenopause: Within the critical window before significant neurodegeneration

• Moderate dosing: Avoiding supraphysiologic estradiol levels, which may be particularly harmful in APOE4 carriers

• Adequate progesterone: To maintain E2:P4 balance and support ERRα function as suggested by the Nature Communications paper

Lifestyle Interventions Offer Evidence-Based Risk Reduction

DHA-enriched fish oil supplementation (at physiologically relevant doses) improved recognition memory in APOE4 females with ovarian failure, increased brain DHA and the DHA:AA ratio, elevated BDNF protein concentration, and mitigated aberrant gene expression related to bioenergetics and blood-brain barrier function.

Physical activity and omega-3 fatty acid intake may modify the impact of APOE4 on Alzheimer’s and cardiovascular disease risk. (see my recent Substack on Omega3s here

https://neurolipidnotebook.substack.com/p/estrogen-dha-and-apoe4-interactions)

Multidomain lifestyle interventions incorporating diet, exercise, cognitive training, and vascular risk monitoring have shown benefits.

Cardiovascular and metabolic health monitoring is essential, as APOE4 is associated with poorer metabolic function and these factors interact synergistically to influence dementia risk. Addressing modifiable risk factors—including physical inactivity, obesity, hypertension, diabetes, and inflammation—may be particularly important in APOE4 carriers.

Monitoring and Individualization

APOE4 carriers considering or using hormone therapy should have close monitoring of functional outcomes (cognition, sleep, mood, vasomotor symptom control) rather than relying solely on hormone levels. These functional responses may be better indicators of appropriate hormonal balance than serum hormone levels alone.

The decision to use hormone therapy should involve extensive shared decision-making considering individual risk factors, family history, APOE4 status, symptom burden, and personal preferences.

---

Take-Home Message

This study doesn’t tell us exactly what to prescribe or what lab values are “ideal.” But it does give us a much clearer brain-centric biological narrative:

It’s the balance of estradiol to progesterone—and the formulation of that progesterone—not just the amount of either, that matters for neural energy regulation and potential Alzheimer’s risk.

That opens the door to:

• Personalized hormone therapy strategies emphasizing bioidentical formulations

• Monitoring functional responses (sleep, cognition, mood)

• Integrating metabolic support alongside hormone management

• Recognizing APOE4 carriers as a population requiring particularly careful consideration of timing, formulation, and balance

---

Looking Ahead

Future research should aim to define:

• Functional ranges for E2:P4 that support ERRα signaling…

• How different delivery methods (transdermal vs oral) and formulations (bioidentical vs synthetic) influence brain hormone ratios

• The timing of hormone interventions relative to menopause

• Interactions with genetics (e.g., APOE4)

• Prospective RCTs specifically recruiting APOE4-positive perimenopausal women comparing bioidentical hormone formulations versus placebo

Until then, thinking in ratios, formulations, and functional biology rather than single hormone levels can be a more productive framework for clinicians and curious patients alike—particularly for APOE4 carriers who may represent both the highest-risk and potentially highest-benefit population for appropriately timed and formulated hormone therapy.

Personally, this study gives me great pause in thinking about my own perimenopausal care a a APOE4 carrier. I know that the SYNTHETIC progesterone released in the IUD I use to protect my uterus from unopposed estrogen is systemic (can be traced to breast tissue) though whether it penetrates the brain we are unsure, and likely offers little to no central nervous system buffering of estrogen signaling in my brain. When I first started estrogen therapy my provider followed standard guidance and said that because I had an IUD, she would NOT recommend systemic progesterone because it was not needed. This may be a prime example about what contemporary women’s health experts call old fashioned Bikini Medicine. This is the concept that womens health has long been relegated to the 2 “bikini” areas, the long-standing tendency to focus women’s hormonal care on reproductive organs while neglecting the brain and heart while neglecting the brain and heart. I will continue to follow this unsettled, and at times unsettling, science — particularly as it may be especially relevant for APOE4 carriers like myself.

Has our definition of “adequate progesterone” been too anatomically narrow? Have women been taking synthetic progesterone with no education or appreciation of the differences between synthetic and That’s what this study has made me question very seriously.

Theoretical Considerations for Brain Health

Based on the preclinical evidence discussed earlier in our conversation, there are theoretical reasons why adding systemic bioidentical progesterone to an LNG-IUD plus estrogen regimen might provide additional CNS benefits:

1. Different CNS effects: Levonorgestrel and bioidentical progesterone have fundamentally different effects on allopregnanolone synthesis, hippocampal neurogenesis, and mitochondrial energy metabolism when combined with estradiol

2. E2:P4 ratio in the brain: The Nature Communications paper suggests the brain may require adequate systemic progesterone to maintain E2:P4 balance and ERRα function—whether LNG-IUD-derived systemic levonorgestrel levels provide equivalent CNS protection is unknown

3. APOE4 carriers: Given their baseline mitochondrial dysfunction, APOE4-positive women might particularly benefit from bioidentical progesterone’s support of ATP synthase expression, which levonorgestrel may not provide when combined with estradiol

Safety Considerations

No safety data exist on this combination. Theoretical concerns would include:

• Excessive progestogenic effects (though systemic absorption from LNG-IUD is relatively low)

• Potential for increased side effects (mood changes, breast tenderness)

• Unknown interactions between levonorgestrel and bioidentical progesterone at the receptor level

Clinical Implications

For a perimenopausal woman currently using an LNG-IUD with estrogen therapy who is concerned about brain health (particularly if APOE4-positive), the evidence gap means clinical decisions must be highly individualized. Options to consider include:

1. Continue current LNG-IUD plus estrogen regimen with close monitoring of functional outcomes (cognition, mood, sleep)

2. Switch to systemic bioidentical estradiol plus bioidentical progesterone (removing the IUD), accepting the need for alternative contraception if still needed

3. Experimentally add low-dose systemic bioidentical progesterone to LNG-IUD plus estrogen, recognizing this is entirely unstudied and would require informed consent and careful monitoring

This represents a critical research gap. Future studies should compare: LNG-IUD plus estrogen, systemic bioidentical hormones, and LNG-IUD plus estrogen plus systemic bioidentical progesterone, measuring cognitive outcomes, AD biomarkers, and brain imaging endpoints—particularly in APOE4 carriers.

Until such data exist, the question of whether systemic bioidentical progesterone provides additional CNS protection beyond LNG-IUD in perimenopausal women remains unanswered.

Special thanks to the inspiring research of:

Sun, J.KL., Peng, A.Z., Hart, R.P. et al. Perimenopausal state oestradiol to progesterone imbalance drives Alzheimer’s risk via ERRα dysregulation and energy dyshomeostasis. Nat Commun 16, 11546 (2025). https://doi.org/10.1038/s41467-025-66726-4