The FDA Removed the Breast Cancer Warning From Hormone Therapy. Here's What Actually Changed — and What Didn't.

DobroMedOnline
18 hours ago
17 min read

Updated: 3 hours ago

For twenty-three years, every box of estrogen in an American pharmacy carried a black-bordered warning about heart disease, breast cancer, and dementia.

In February 2026, the FDA started taking most of it off.

Here's what actually happened, and what it means for a real decision in a real exam room.

Why everyone is suddenly talking about this

On November 10, 2025, HHS and the FDA announced they intended to remove the boxed warnings from menopausal hormone therapy. On February 12, 2026, the FDA approved the first six actual label changes — removing the cardiovascular disease, breast cancer, and probable dementia statements from the boxed warning across systemic estrogen, systemic estrogen-plus-progestogen, systemic progestogen, and vaginal estrogen products. Twenty-nine companies had submitted proposed changes; six were approved in that first batch.

Two details matter more than they sound.

First, the box didn't disappear. The endometrial cancer warning stays on estrogen-alone products — the one warning that has always been straightforwardly true, and the reason women with a uterus are prescribed a progestogen alongside estrogen.

Second, labels are still in a mixed state. Pick up a generic estradiol patch this month and the printed insert may well still carry the old warning in full. Nothing about the drug changed.

Here's the part almost no coverage mentioned: this was a labeling decision based on a re-reading of existing evidence. No new clinical trial was published. The same data that supported the warning in 2003 supported removing it in 2026 — because the way we interpret that data has genuinely shifted over twenty years, mostly around who was studied.

What's actually happening in your body

Estrogen doesn't only regulate periods. Receptors for it sit in your brain's temperature-control

center, your blood vessels, your bones, your bladder, and the tissue of the vulva and vagina.

When estrogen production becomes erratic — and in perimenopause it becomes erratic, not simply low — the brain's thermostat gets twitchy. A change in core temperature that used to be ignored now triggers a full heat-dumping response: flush, sweat, heart pounding. That's a hot flash. At 3 a.m. it's a night sweat, and it fragments sleep in a way that shows up the next day as brain fog, irritability, and a shorter fuse than you recognize in yourself.

This is also why the fatigue and mood changes of perimenopause are so often dismissed. They're real, and they're frequently downstream of sleep that's being interrupted eight times a night by a thermostat that has lost its calibration.

The original warning came from a study of the wrong women

The Women's Health Initiative, published in 2002, randomized more than 27,000 women to hormone therapy or placebo. It remains the largest randomized trial of menopausal hormone therapy ever conducted. But the average participant was 63 years old and more than a decade past her final period. Only about a third were in their 50s.

That matters, because the results split sharply by age and by time since menopause. Looking at coronary heart disease by how long a woman had been postmenopausal:

(Hazard Ratio — a statistical measure comparing how often an event occurs over time between two groups)

Less than 10 years since menopause: HR 0.76 (95% CI 0.50–1.16)
10 to 19 years: HR 1.10 (0.84–1.45)
20 or more years: HR 1.28 (1.03–1.58) — a real increase

The trend across those groups was statistically significant (P = 0.02).

Starting hormones at 52 and starting them at 72 are not the same intervention.

Breast cancer: the number people actually want

In the WHI arm using conjugated estrogens plus medroxyprogesterone (a synthetic progestin), invasive breast cancer risk rose: HR 1.24 (1.01–1.53), which works out to about 9 extra cases per 10,000 women per year. Over 20 years of follow-up the hazard ratio was 1.28.

In the arm using estrogen alone — women who'd had a hysterectomy — breast cancer risk went the other direction: HR 0.78 (0.65–0.93) at 20 years, with breast cancer mortality also lower (HR 0.60, 0.37–0.97).

Read that again. Estrogen alone did not increase breast cancer in the largest randomized trial of hormone therapy ever conducted — it reduced it. The increase in the combined arm appears to be driven substantially by the progestin.

We should say plainly that this is not unanimous. In the French E3N cohort discussed below, estrogen alone was associated with a modest increase (RR 1.29, 1.02–1.65). Randomized data and observational data disagree here, and that disagreement has not been resolved.

The most useful way to hold the number is comparative. For 1,000 women aged 50–59 over five years, against a background rate of roughly 23 breast cancers:

Combined estrogen + progestogen therapy — +4
Estrogen-only therapy — −4
Two or more units of alcohol daily (UK units — roughly one US standard drink or more) — +5
BMI of 30 or higher — +24
Moderate exercise of 2.5 hours or more weekly — −7

(British Menopause Society / NICE. The BMS notes this infographic is under statistical review following the 2024 NICE update.)

We want to be honest about the other side of this. A very large observational meta-analysis published in The Lancet in 2019 produced higher estimates — roughly 1 extra breast cancer per 50 women who used combined therapy for 5 years, counted over a 20-year horizon. That analysis has real limitations: about 40% of the data came from a single screened cohort, it was entirely observational, and it used older synthetic progestins almost exclusively. It also directly contradicts the randomized WHI result for estrogen-only therapy. Both figures are defensible. Quoting whichever one suits your argument is the single most common dishonesty in menopause content, from both directions.

The type and route change the risk

This is where the last two decades genuinely changed practice.

Route matters for clots. In a study of over 80,000 venous thromboembolism cases (BMJ, 2019), oral hormone therapy carried an odds ratio of 1.58 — about 9 extra clots per 10,000 women per year. Transdermal estradiol — patches, gels, sprays — carried an odds ratio of 0.93. No detectable increase. Estrogen absorbed through skin bypasses the liver, and it's the liver's first pass that drives clotting factor production.

Route matters for stroke, with a dose caveat. Low-dose transdermal estrogen at 50 mcg or less showed no increase in stroke (RR 0.81, 0.62–1.05). Above 50 mcg, the risk rose (RR 1.89, 1.15–3.11). "Patches are safe" is close to true, and quietly dose-dependent.

The progestogen matters. In the French E3N cohort of over 80,000 women, estrogen combined with micronized progesterone — the bioidentical form, sold as Prometrium and available as an inexpensive generic — showed a relative risk of 1.00 (0.83–1.22) for breast cancer. Estrogen with other synthetic progestogens: 1.69 (1.50–1.91). (In the same study, estrogen alone was 1.29 — the figure noted above.) This is observational data, and the reassurance is best supported through about five years; beyond that, the evidence thins out considerably.

Dementia: what we know and what we don't

In WHI participants aged 65 and older, combined therapy roughly doubled probable dementia (HR 2.05, 1.21–3.48). That finding is real and is the reason initiating hormone therapy in your 70s is not recommended.

In women who started near menopause, the picture is different — and empty. The KEEPS Continuation study followed women about ten years after four years of hormone therapy started in early menopause and found no cognitive difference in either direction. A 2025 systematic review and meta-analysis of just over a million participants found no significant association between hormone therapy use and risk of mild cognitive impairment or dementia, and concluded that hormone therapy should be prescribed on its other benefits and risks, "not for dementia prevention."

So: hormone therapy is not a dementia prevention strategy. It is also not established as a dementia cause when started early. Anyone telling you either with confidence is ahead of the data.

Do you actually need labs?

Mostly, no — and this surprises people.

If you are over 45 with typical symptoms, menopause is a clinical diagnosis. NICE guidance is explicit: don't use FSH, estradiol, AMH, inhibin, or antral follicle count to diagnose perimenopause or menopause in women over 45.

The reason is mechanical. In perimenopause, FSH and estradiol swing wildly week to week. A "normal" result doesn't rule anything out and an "elevated" one doesn't confirm anything. The direct-to-consumer "menopause panel" is measuring a moving target and selling you the snapshot.

Testing genuinely earns its place when:

You're under 40 and periods have stopped or become irregular — this is a workup for primary ovarian insufficiency, and it changes everything about management
You're 40–45 with menopausal symptoms and menstrual change
You've had a hysterectomy or endometrial ablation, so the 12-month rule can't be applied
Symptoms suggest something else entirely — thyroid disease, anemia from heavy perimenopausal bleeding, depression — and you need to sort the differential

Worth doing at this age regardless of hormones: blood pressure, a lipid panel, an A1c, and — under the 2026 ACC/AHA dyslipidemia guideline — a once-in-a-lifetime Lp(a). The menopause transition itself raises LDL and ApoB independent of aging, and the new guideline explicitly names early menopause before 45 and premature menopause before 40 as cardiovascular risk enhancers. The perimenopausal visit is the best cardiovascular prevention opportunity most women get.

What we don't recommend: salivary or urine hormone panels to guide dosing. Every major society has rejected them — after topical progesterone, salivary levels have been measured across a more than 300-fold range. There is no dose you can rationally set from that number.

"Bioidentical" and "compounded" are not the same word

This confusion costs patients real money.

Estradiol is bioidentical. It is molecularly identical to the estrogen your ovaries made. It is FDA-approved, comes as patches, gels, sprays, oral tablets, and vaginal preparations, and most forms have inexpensive generics.

Micronized progesterone is bioidentical. FDA-approved, generic, inexpensive.

Neither requires a compounding pharmacy. When someone is told they need "compounded bioidentical hormones" for a natural option, they are usually being sold, at cash price, a less regulated version of a drug their insurance would have covered.

The 2020 National Academies report reviewed the compounded market and found only 13 studies of adequate rigor across the entire literature, concluding there was "insufficient evidence to support the overall clinical utility" of compounded hormone therapy. Compounding is genuinely reasonable in a narrow set of situations — a documented allergy to a component of an approved product (peanut oil in some progesterone capsules is the classic), a dose or route that simply isn't manufactured, or testosterone for women, where the US has no FDA-approved female product at all.

Pellets are a separate conversation, and our position is firmer.

They produce hormone levels that are unpredictable and often far above the physiologic range, and — this is the core problem — once implanted, the dose cannot be lowered or stopped. ACOG explicitly recommends against pellets for testosterone delivery, citing the inability to remove them. The Global Consensus Position Statement on testosterone in women recommends against any preparation producing supraphysiologic levels, "including pellets and injections."

And what about testosterone?

It comes up in almost every conversation now, so here is the honest version.

There is no FDA-approved testosterone product for women in the United States. Every use here is off-label — usually a small fraction of a gel approved for men. That gap is real, and it is exactly the vacuum that pellet clinics and compounding pharmacies have moved into.

The only evidence-based use — per an international consensus statement endorsed by The Menopause Society, the Endocrine Society, ACOG, ISSWSH and others — is distressing low sexual desire in postmenopausal women, after a proper assessment.

How big is the effect? The largest analysis pooled 36 randomized trials and 8,480 women and found roughly one additional satisfying sexual encounter per month compared with placebo, along with improvements in desire, arousal, orgasm, and sexual self-image. That is a real benefit and it matters to some women a great deal. It is also nothing like what the marketing implies, and you deserve both halves of that sentence.

What it does not do, according to that same consensus: there is "no effect of testosterone therapy on general wellbeing," no effect on depressed mood, insufficient evidence for cognition, and "the available data do not support an effect of testosterone treatment on bone mineral density." Energy, brain fog, muscle, and vitality are the reasons it is most often sold and the ones with the least support behind them.

And no blood test can tell you your testosterone is low. The consensus statement is unusually blunt: "No cutoff blood level can be used for any measured circulating androgen to differentiate women with and without sexual dysfunction." There is no such diagnosis in women. Levels are checked after you start — to make sure you aren't getting too much — not to decide that you need it. A clinic that draws a level, shows you a number, and sells you a treatment from it is doing something the entire international consensus says cannot be done.

If it is used: a transdermal cream or gel at roughly a tenth of a male dose, a level rechecked at 3–6 weeks and then twice yearly, and a firm stopping rule — if six months brings no meaningful benefit, stop. Not oral, not injected, not pellets; all three overshoot the physiologic range. And be clear-eyed about the horizon: trials run to about two years and look reassuring for breast density and cardiometabolic markers. Beyond that, the safety data thin out.

What about supplements?

We partner with Fullscript dispensary, so we have every commercial reason to soften this. We won't.

For hot flashes specifically, the evidence does not support supplements. The Menopause Society's 2023 position statement reviewed them and classified supplements and herbal remedies as not recommended, including black cohosh (which failed against placebo in a Cochrane review of 16 trials and carries a documented liver-injury signal), soy and equol, evening primrose, maca, dong quai, and pollen extract. Vitamin E beat placebo by exactly one hot flash per day — statistically significant, clinically meaningless; women in the trial couldn't tell which arm they were in.

What the same statement does recommend for hot flashes: cognitive behavioral therapy, clinical hypnosis, SSRIs and SNRIs, gabapentin, oxybutynin, fezolinetant, and weight loss.

So where do supplements legitimately fit in midlife? Not as hot flash treatment. They fit where the goal is different:

Vitamin D and calcium work as a floor, not a lever. Correcting a deficiency matters, and adequacy is a precondition for every osteoporosis drug to work. Taking more once you're replete does not reduce fractures — a 2,000 IU per day trial in nearly 26,000 adults found no fracture benefit.
Magnesium has thin, low-certainty evidence for falling asleep faster, about 17 minutes in a three-trial meta-analysis. Cheap, low-risk, worth a trial. Not a proven therapy — and separate it by four hours from levothyroxine and bisphosphonates.
Protein and creatine are the honest muscle story. Aim for roughly 1.0–1.2 g/kg/day of protein, higher if you're strength training or losing weight. Creatine at 3–5 g/day plus resistance training increased lean mass in a two-year randomized trial of 237 postmenopausal women — and showed no benefit for bone density, which is the opposite of what most online content claims.
Pregnenolone deserves a specific mention because it's sold as a menopause and "adrenal" supplement and it is not a vitamin — it's a genuine steroid hormone, the precursor to progesterone, DHEA, testosterone, estrogens, and cortisol. No trial supports it for menopausal symptoms. The best controlled trial of it was in chronic back pain, and it found no improvement in sleep, mood, memory, or executive function — exactly what it's marketed for. It also can't be monitored: there's no target level, and you can't steer where an upstream precursor ends up, because the enzymes below it are tissue-specific. And the "pregnenolone steal" story you may have read — that stress diverts pregnenolone away from your sex hormones to make cortisol — isn't how steroid physiology works. Those hormones are made in different tissues with different enzymes; there's no shared pool to drain.
Ashwagandha is the one we'd ask you to be most careful with. It is the most aggressively marketed supplement to women your age, and it carries a documented liver-injury signal plus reports of thyroid disturbance — a genuine problem if you're one of the many midlife women on levothyroxine.

One honest caveat, because we just criticized selective quoting: that same position statement also lists exercise as not recommended specifically for hot flashes — trials have not shown it reduces flash frequency. Resistance training two to three times a week is still the single highest-value thing in this article for muscle, bone, metabolic health, and function after 50. It is just not a hot flash treatment, and we're not going to pretend it is. Nobody sells it either way.

How long before you feel a difference?

This question decides more outcomes than almost anything else, because most treatments that "didn't work" were abandoned at week three.

Two things happen early and they pull opposite ways: the benefit builds slowly and the side effects arrive fast. Breast tenderness in week two feels like proof this was a mistake. It usually isn't — it's the most common early effect of estrogen and it typically settles by 6 to 12 weeks.

Estrogen (patch, gel, spray, or pill): something noticeable in 1–2 weeks, often night sweats first. Judge it properly at 8–12 weeks.
Vaginal estrogen: 2–3 weeks for symptoms, 8–12 weeks for the real answer, and up to 6 months if the tissue changes are advanced.
SSRIs or SNRIs for hot flashes: 1–2 weeks — faster than when those drugs are used for depression.
Gabapentin: 1–2 weeks.
Fezolinetant or elinzanetant: judge at week 12. No dose to adjust; if it hasn't worked by then, it won't.
Testosterone: 3–6 months, with a firm stop at six if nothing has changed.

The standard review point is 3 months — guidelines recommend checking at three months for both effect and tolerability, then annually.

And when should the dose go up? Not before 4–6 weeks, because before that you're judging something that hasn't finished arriving. The usual rhythm is: review at 6–8 weeks, and if symptoms are still bothersome and you're tolerating it, go up one step, not two — then reassess 6–8 weeks later. Most women settle in the middle of the range, not at the top.

Before the dose goes up, five things are worth checking, because they explain most apparent failures:

is the patch staying on and the gel going on dry skin;
would a different delivery form absorb better;
is the leftover symptom actually vaginal or urinary, in which case more systemic estrogen is the wrong lever;
is something else going on — thyroid, iron, sleep apnea, mood;
and is your sleep being broken by something other than night sweats.

One honest caveat: transdermal estrogen up to 50 mcg showed no increase in stroke risk, and above 50 mcg it did. That doesn't make higher doses wrong — for someone whose life is being wrecked by symptoms it's often exactly right. It makes it a decision you should be part of, rather than something that drifts upward.

Who should pay attention?

Hormone therapy is most favorable if you are under 60 or within 10 years of your final period, have bothersome symptoms, and without contraindications.

Systemic estrogen is contraindicated with undiagnosed vaginal bleeding, current or prior breast cancer, estrogen-dependent cancer, active or prior DVT or PE, active or prior stroke or heart attack, liver disease, or a known thrombophilia.

A special case that gets missed: if your periods stopped before 40 (primary ovarian insufficiency) or before 45 (early menopause), hormone therapy is not a lifestyle choice. It's replacement of something you're supposed to have, usually continued until around the natural age of menopause, and it protects bone and cardiovascular health.

Vaginal dryness and painful sex are a different problem with a different answer. Low-dose vaginal estrogen is minimally absorbed — serum levels stay in the postmenopausal range — requires no progestogen, and treats what systemic therapy often doesn't fully fix. In women with a history of breast cancer, ACOG's 2021 consensus says low-dose vaginal estrogen may be used after nonhormonal options have failed, including in women on tamoxifen, and with shared decision-making involving the oncologist for those on aromatase inhibitors.

What you can do

Track what's actually happening for two weeks before your visit: how many hot flashes a day, how many wake you up, what your sleep looks like, and — this is the question that gets skipped — what it's costing you at work and at home.
Bring the real list: every medication, every supplement, and your family history of breast cancer, blood clots, and heart disease.
Ask about route, not just "hormones or not." For most women with any clot or cardiovascular concern, transdermal estradiol plus micronized progesterone is the better-supported starting point.
Get the cardiovascular numbers — blood pressure, lipids, A1c, Lp(a) once. Do it whether or not you take hormones.
Start resistance training. Two to three sessions a week. Not for the hot flashes — for the muscle, bone, and metabolic health that determine how the next thirty years go.

When to talk to a healthcare provider

Sooner rather than later if you have:

Any bleeding after 12 months without a period. This always requires evaluation.
New bleeding on hormone therapy starting more than 6 months after you began, or more than 3 months after a dose change
Periods that have stopped before age 40
Chest pain, one-sided leg swelling, sudden shortness of breath, or new weakness, numbness, or difficulty speaking — stop and seek emergency care
Hot flashes with unexplained weight loss, drenching night sweats without flushing, or fever — these deserve a broader workup than menopause

The DobroMed bottom line

The FDA's label change corrected a warning that had been applied too broadly for too long — especially to vaginal estrogen, which was never the drug the warning was about. It did not make hormone therapy risk-free, and no new trial has been published.

For a healthy woman under 60 or within ten years of her last period, with bothersome symptoms and no contraindications, the benefits of hormone therapy generally outweigh the risks. Transdermal estradiol with micronized progesterone is the best-supported combination available today, and every component of it is FDA-approved, bioidentical, and usually generic.

For hot flashes, supplements have been studied and have not performed. We'd rather tell you that than sell you something.

And you almost certainly don't need a hormone panel to make any of these decisions.

PATIENT SELF-ASSESSMENT

Most women don't get a clear answer about perimenopause. They get "your labs are normal," or "you're too young for that," or a list of symptoms that fits five different conditions equally well.

This questionnaire won't diagnose you — nothing that lives on a webpage can. What it will do is turn a vague sense that something has changed into something specific: where you likely are in the transition, how heavy your symptom load actually is, which symptoms are driving it, and what the evidence supports for your particular pattern.

It takes about eight minutes. Section B is the Menopause Rating Scale, a validated instrument used in clinics and research worldwide — we kept it whole, because shortening it would break its scoring. Everything after it is short, and only appears if something needs a follow-up question.

The summary at the end is meant to leave with you. Bring it to your appointment; it's built so a clinician can read it in under a minute.

Have you ever been told you needed a hormone level checked to confirm menopause — or been recommended compounded pellets? Tell us in the comments. We're building the next article from what people are actually being told.

Sources

FDA. FDA Approves Labeling Changes for Menopausal Hormone Therapy Products. February 12, 2026.
The Menopause Society. Comments on the FDA Announcement on Hormone Therapy. November 2025.
The Menopause Society. The 2022 Hormone Therapy Position Statement. Menopause. 2022;29(7):767–794.
The Menopause Society. The 2023 Nonhormone Therapy Position Statement. Menopause. 2023;30(6):573–590.
Manson JE, Crandall CJ, Rossouw JE, et al. The Women's Health Initiative Randomized Trials and Clinical Practice: A Review. JAMA. 2024;331(20):1748–1760.
Chlebowski RT, Anderson GL, Aragaki AK, et al. Association of Menopausal Hormone Therapy With Breast Cancer Incidence and Mortality During Long-term Follow-up of the WHI Randomized Clinical Trials. JAMA. 2020;324(4):369–380.
Rossouw JE, Prentice RL, Manson JE, et al. Postmenopausal Hormone Therapy and Risk of Cardiovascular Disease by Age and Years Since Menopause. JAMA. 2007;297(13):1465–1477.
Collaborative Group on Hormonal Factors in Breast Cancer. Type and timing of menopausal hormone therapy and breast cancer risk. Lancet. 2019;394(10204):1159–1168.
Fournier A, Berrino F, Clavel-Chapelon F. Unequal risks for breast cancer associated with different hormone replacement therapies. Breast Cancer Res Treat. 2008;107(1):103–111.
Vinogradova Y, Coupland C, Hippisley-Cox J. Use of hormone replacement therapy and risk of venous thromboembolism. BMJ. 2019;364:k4810.
Renoux C, Dell'Aniello S, Garbe E, Suissa S. Transdermal and oral hormone replacement therapy and the risk of stroke. BMJ. 2010;340:c2519.
Gleason CE, Dowling NM, Kara F, et al. Long-term cognitive effects of menopausal hormone therapy: KEEPS Continuation. PLOS Med. 2024;21(11):e1004435.
Melville M, He L, Desai R, et al. Menopause hormone therapy and risk of mild cognitive impairment or dementia: systematic review and meta-analysis. Lancet Healthy Longev. 2025;6(12):100803.
Davis SR, Baber R, Panay N, et al. Global Consensus Position Statement on the Use of Testosterone Therapy for Women. J Clin Endocrinol Metab. 2019;104(10):4660–4666.
Islam RM, Bell RJ, Green S, Page MJ, Davis SR. Safety and efficacy of testosterone for women: a systematic review and meta-analysis of randomised controlled trial data. Lancet Diabetes Endocrinol. 2019;7(10):754–766.
National Academies of Sciences, Engineering, and Medicine. The Clinical Utility of Compounded Bioidentical Hormone Therapy. 2020.
ACOG. Compounded Bioidentical Menopausal Hormone Therapy. Clinical Consensus No. 6. Obstet Gynecol. 2023.
ACOG. Treatment of Urogenital Symptoms in Individuals With a History of Estrogen-dependent Breast Cancer. Clinical Consensus No. 2. 2021.
NICE. Menopause: identification and management. NG23. Updated November 2024.
2026 ACC/AHA Multisociety Guideline on the Management of Dyslipidemia. J Am Coll Cardiol. 2026;87(19):2624–2757.
Leach MJ, Moore V. Black cohosh (Cimicifuga spp.) for menopausal symptoms. Cochrane Database Syst Rev. 2012;(9):CD007244.
Chilibeck PD, et al. A 2-Year Randomized Controlled Trial on Creatine Supplementation During Exercise for Postmenopausal Bone Health. Med Sci Sports Exerc. 2023.
LiverTox: Clinical and Research Information on Drug-Induced Liver Injury. Ashwagandha. NIDDK.

This article is educational and is not a substitute for individual medical advice. Hormone therapy decisions depend on your personal and family history. If you'd like to review whether hormone therapy is appropriate for you, DobroMed Online can go through your history, symptoms, and risk factors and help you decide what — if anything — is worth doing.