Overview of #381 ‒ Alzheimer's disease in women
This episode of The Drive (host Peter Attia, MD) features Lisa Mosconi, Ph.D., neuroscientist, neuroimager, and director of the Women’s Brain Initiative at Weill Cornell. The conversation centers on why Alzheimer’s disease (AD) disproportionately affects women, how hormonal transitions (puberty, pregnancy/postpartum, perimenopause/menopause) influence brain aging and AD risk, what modern imaging and blood biomarkers reveal about preclinical disease, the complex evidence around menopausal hormone therapy (MHT/HRT), genetics (notably APOE4), lifestyle and therapeutics (GLP-1s, SERMs/CIRMs), and Mosconi’s CARE Initiative — a $50M Wellcome Leap-funded program to cut women’s AD risk in half by 2050.
Key takeaways
- Alzheimer’s disproportionately affects women (~2:1 prevalence). Longevity alone does not explain that gap; sex-specific biology appears central.
- AD is fundamentally a midlife disease: pathological processes begin decades before clinical symptoms; cognitive tests are often insensitive to early / subjective changes.
- Menopausal transition is a critical neurobiological event — brain energy metabolism, structure, and immune signaling change during perimenopause/menopause.
- Hormone dynamics matter: brain estrogen receptor density and function (not just blood estradiol) likely determine neural effects; human data don’t fully match animal models.
- Timing and formulation of MHT appear crucial (the “timing hypothesis”): benefits likely greater when started near menopause; WHI results (oral CEE + MPA) caused long-standing confusion because of age, formulation, and population issues.
- APOE4 confers a larger relative AD risk in women than in men (heterozygous women ≈ 4× risk; homozygous ≈ 12–15×).
- Robust lifestyle interventions (sleep, exercise, diet, cardiometabolic risk control) remain the primary, accessible prevention tools.
- CARE Initiative: three-year, high-speed research sprint with multiple sites and global datasets to better define neuroendocrine contributions and actionable strategies.
The biological and epidemiologic picture
- Dementia is an umbrella term; AD is the most common (~70%). Other dementias (frontotemporal, Lewy body, vascular) have different pathologies and mixed presentations are common.
- Preclinical AD can last decades: molecular lesions (amyloid, tau), metabolic and vascular changes precede measurable cognitive deficits.
- Women show more midlife “red flags” on imaging and faster lesion progression, but often maintain better performance on verbal memory tests (cognitive reserve), delaying clinical diagnosis despite greater pathology.
- The question shifts from “Are women living longer?” to “What happens to female brains in midlife that increases lifetime AD risk?”
Imaging and biomarker tools discussed
- Structural MRI (T1, T2/FLAIR): volumetrics (hippocampus, medial temporal lobe), white matter hyperintensities, gliosis.
- DTI (diffusion tensor imaging): white-matter structural connectivity.
- ASL (arterial spin labeling MRI): cerebral blood flow.
- 31P MR spectroscopy: phosphorus spectroscopy to measure brain energetics (ATP / phosphocreatine ratios).
- FDG-PET: cerebral glucose metabolism (functional).
- Amyloid PET (PiB, C11-PiB or fluorinated tracers): amyloid plaque imaging.
- New brain estrogen imaging: F-18 fluroestradiol (FES) PET measures estrogen receptor (ER) density (so far validated in pituitary; tracer is currently more alpha-selective and originally developed for oncology). Typical imaging uptake peaks ~30–50 minutes; radiation exposure is low (<1 mSv).
- Blood-based biomarkers (p-tau, amyloid, neurofilament light, inflammatory markers): promising, minimally invasive tools for tracking disease biology and response to interventions; useful for research and intervention monitoring.
Estrogen, receptors, and menopause — new human imaging findings
- Mosconi’s group published the first human imaging comparing pre-, peri-, and postmenopausal brains. Key findings:
- Pre-menopause, male/female brain differences are minimal; changes emerge across the perimenopausal transition.
- Using FES-PET, ER density in the pituitary increases during perimenopause and is higher postmenopause up to age ~65 — contrary to many rodent ovary-removal models that predict a post-menopausal ER decline.
- Human brain compensation: as circulating estradiol falls, the brain may upregulate ER density to “grab” available hormone; when the system eventually collapses the receptors may decline, but that appears to happen later (>65) in Mosconi’s cohort.
- Important caveat: current FES tracers are more ER-α selective, and receptor presence doesn’t directly prove downstream transcriptional function. Functional receptor integrity and signaling remain to be established.
- Implication: the “window of opportunity” for MHT effects on the brain may be broader than rodent models suggested; human receptor dynamics are different.
Hormone therapy (MHT/HRT): evidence, timing, and nuances
- The Women’s Health Initiative (WHI) memory study used oral conjugated equine estrogen (CEE) ± medroxyprogesterone acetate (MPA) in older postmenopausal women and reported increased dementia risk — results that created decades of caution and confusion.
- Key points from current evidence:
- Clinical trial evidence on dementia incidence is limited (WHI is the only major RCT with dementia endpoints).
- Observational meta-analyses suggest timing matters:
- Estrogen-only therapy started within ~10 years of final menstrual period in women with prior hysterectomy: ~32% reduced dementia risk in pooled observational analyses.
- Estrogen + progesterone started within ~10 years (women with uterus): trend toward ~23% reduced risk (mixed results across studies).
- Initiation >10 years after menopause showed no benefit and possible increased risk for combined therapy.
- Formulation matters: oral CEE + MPA differs biologically from transdermal estradiol + micronized progesterone; MPA (the WHI progestin) implicated in vascular and breast concerns; transdermal estradiol has different effects on coagulation and vascular risk.
- Observational studies are subject to healthy-user and indication biases; more mechanistic and biomarker-guided prospective work is needed.
- Practical takeaway Mosconi emphasizes: perimenopause is the right time for conversation/intervention; individualized, precision approaches matter (dose, route, type of progestogen, symptom profile, risk factors).
Genetics — APOE4 and sex interactions
- APOE4 increases AD risk more strongly in women:
- Heterozygous APOE4 (one allele) in women ≈ 4× dementia risk vs non-carriers.
- Homozygous APOE4 (two alleles) in women ≈ 12–15× risk.
- Women carrying APOE4 tend to develop more pathology earlier and to progress faster.
- Understanding sex-by-genotype interactions is a priority for targeted prevention and treatment.
Therapeutics under consideration (GLP-1s, CIRMs/NeuroSERMs, and others)
- GLP-1/GIP dual agonists (e.g., tirzepatide) and GLP-1s:
- Biological plausibility: improve insulin sensitivity, reduce inflammation, influence neuroinflammation; GLP-1 receptors are present in the brain.
- Early/pilot data (small N) suggest reductions in blood/CSF markers of neuroinflammation and aggregation; brain-specific tracers for GLP-1 are in development.
- Large-scale, biomarker-driven RCTs are needed to test cognitive or disease-modifying effects independent of weight/metabolic changes.
- Selective estrogen receptor modulators (SERMs) and neuroSERMs (CIRMs):
- NeuroSERMs can be designed to target brain ER-β (cognitive regions) while sparing reproductive tissues; Roberta Diaz-Brinton’s phyto-derived neuroSERM (a GRAS supplement) is in phase 2 trials for cognition and brain energetics.
- CIRMs offer an attractive strategy to get brain benefit without peripheral risks, but clinical outcome data are pending.
- Other agents and approaches — Clotho, amyloid/tau-targeting biologics, anti-inflammatory strategies — are moving forward; optimism has grown over the last 5–10 years relative to a decade ago.
CARE Initiative — Cutting Alzheimer’s Risk through Endocrinology
- Funded by Wellcome Leap: $50M, three-year “sprint” to rapidly generate high-impact data on women’s brain health and hormonal aging.
- Goals:
- Better define neuroendocrine aging and how reproductive history (puberty, pregnancy, postpartum, perimenopause/menopause) influences AD risk.
- Use imaging, blood biomarkers, and longitudinal cohorts to map risk and test interventions.
- Leverage large global datasets (CARE aims to access female-specific data across six continents and >20 million women’s data) and multi-site biomarker studies.
- Run focused, rapid-impact studies (including observational cohorts of women who elect to start MHT versus not, and biomarker endpoints) to inform RCT design and clinical guidance.
- Target: halve women’s AD risk by 2050 (estimates of preventing tens of millions of cases globally if successful).
Practical recommendations for women (and clinicians)
- Lifestyle remains foundational:
- Prioritize sleep, regular physical activity (consistent, long-term exercise to support BDNF and neuroplasticity; moderate-to-vigorous intensity tailored to time availability), cardiometabolic health (blood pressure, glucose/insulin control), balanced diet, stress reduction.
- Behavioral change needs consistency — brain resilience builds slowly but is durable.
- Discuss menopause proactively:
- Perimenopause is a key window to evaluate symptoms and risk; MHT should be part of informed, individualized discussion (timing, route, formulation, personal / family risk factors).
- Transdermal estradiol + micronized progesterone is commonly used today and has different risk profiles than oral CEE + MPA used in WHI.
- Biomarkers and specialized assessment:
- Blood-based biomarkers (p-tau, amyloid, NfL) and advanced imaging are increasingly useful for research and for tracking interventions; specialized Alzheimer prevention clinics can provide access to these measures.
- If concerned about cognitive decline or AD risk (family history, APOE4, strong midlife symptoms), consider evaluation at a research/academic center that can provide biomarker-based monitoring and trial opportunities.
- If you had an early oophorectomy or surgically induced menopause, current clinical guidance generally supports MHT until approximate natural menopause age; do not automatically avoid hormone therapy on the basis of weak observational studies.
- For women considering MHT solely for brain health: evidence is evolving; current data suggest potential benefit when started near menopause, but type and individual risks matter — discuss with clinicians familiar with menopause neurology.
Notable quotes / distilled insights
- “Alzheimer’s is not a disease of old age — it’s a disease of midlife with symptoms that manifest later.”
- “The brain is built for stability; the rest of the body is built for change. Building brain resilience takes time and consistency.”
- Human brain estrogen receptor dynamics differ from rodent ovary-removal models — imaging shows ER density may remain high postmenopause up to ~65, challenging assumptions about a narrow “window.”
Research gaps and needs
- Precise mapping of the menopause “window of opportunity” for MHT: better human biomarker evidence (ER imaging, downstream receptor function, mitochondrial/oxidative stress measures).
- RCTs of MHT with biomarker endpoints (not feasible for 20–30-year dementia outcomes, but feasible for surrogate markers).
- Clarify effects of specific progestogens (MPA vs micronized progesterone) and routes (oral vs transdermal) on brain/vascular outcomes.
- Large, international, sex-stratified analyses of risk factors (incidence vs prevalence) and genetics (sex × APOE interactions).
- Trials testing GLP-1s, CIRMs, and other candidate therapeutics for neuroprotection with biomarker-driven endpoints.
Action items (for clinicians, researchers, and patients)
- Clinicians: integrate menopause history and reproductive factors into cognitive risk assessments; consider biomarker-based referral when appropriate; individualize MHT discussions.
- Researchers: design biomarker-guided, sex-specific, and timing-sensitive studies; leverage global datasets and multidisciplinary collaborations (examples: CARE).
- Patients (women in midlife): optimize lifestyle risk factors, have informed discussions about perimenopause and MHT with clinicians familiar with current evidence, consider participation in research/biomarker programs if concerned.
For more detail, listen to the full interview (Peter Attia with Lisa Mosconi) and check the show notes at peteratiamd.com/show-notes.