Just the FactsThe conclusion in one minute
Older research was sometimes applied too broadly. That does not mean bioidentical hormones carry no breast cancer risk.
Effective does not mean risk-free: hormone therapy can greatly relieve troublesome menopausal symptoms. Some treatments increase breast cancer risk; that does not automatically make treatment unwise.
- WHI has not been debunked. It studied specific hormone tablets in generally older women. Its findings do not automatically apply to every current treatment.
- Bioidentical describes molecular structure, not safety. Estradiol and progesterone can still have side effects.
- Breast cancer risk may be lower with progesterone than with some other progestogens. This does not prove that estradiol plus progesterone adds no risk, especially with long-term use.
- For many healthy women with troublesome symptoms, the balance may be favourable before age 60 or within 10 years of menopause. The specific treatment and personal risk factors remain decisive.
General information, not personal treatment advice. Discuss medication, dose, medical history and duration with your clinician. Do not change treatment based on this document alone.
1. The most important distinction first
Bioidentical means that a hormone has the same molecular structure as a hormone the body produces itself. In menopausal therapy, this mainly refers to 17-beta-estradiol and progesterone. “Micronised” progesterone consists of very small particles of progesterone, allowing it to be absorbed more readily as a medicine. The term says nothing about a proven absence of cancer risk.[6]
Licensed medicines
Examples include estradiol as a patch, gel, spray or tablet, and micronised progesterone as a capsule. These can therefore be standard, routinely prescribed bioidentical hormones. Their composition, manufacture and product information are subject to medicines regulation.
Custom-made preparations
Known as compounded bioidentical hormones: for example, a specially formulated hormone cream or an implant. The same evidence on dosage, absorption, effectiveness and safety of the finished product is not automatically available for these preparations.
There may be a legitimate medical reason for a pharmacy-compounded preparation, such as an allergy to an inactive ingredient. But “custom-made” is not in itself evidence that a medicine is safer or more effective. ACOG, the US professional association of obstetricians and gynaecologists, advises against routinely using such preparations when a suitable licensed treatment is available. US regulations are not the same as those in the Netherlands; however, the lack of evidence for superior safety remains relevant.[7]
Also worth knowing: progesterone is one of the progestogens, the collective term for substances with progesterone-like effects. Other progestogens include medroxyprogesterone acetate (MPA), dydrogesterone and levonorgestrel. Dydrogesterone is not bioidentical, even though some of its safety data are relatively favourable. “Bioidentical versus synthetic” is therefore too simple a distinction; bioidentical medicines are also manufactured industrially.[5][6]
2. What really happened with the WHI research?
“That study” usually refers to the Women's Health Initiative (WHI), particularly the 2002 publication. This involved two large US trials in which women were randomly assigned to receive hormones or a placebo — a pill with no active ingredient. The women were aged 50 to 79, with an average age of about 63. The central question was the balance of benefits and risks in preventing chronic diseases, not just treating hot flushes around the final menstrual period.[1][2]
Trial A: combined therapy
16,608 women with a uterus. They received daily tablets containing a mixture of oestrogens, abbreviated as CEE, plus the progestogen MPA, or a placebo.
This combination was not estradiol plus micronised progesterone.
Trial B: oestrogen alone
10,739 women without a uterus. They received CEE without an added progestogen, or a placebo.
This was a different group of women, not a direct randomised comparison of “with or without progesterone”.
In 2002, the combined-therapy trial was stopped early: there were more breast cancer diagnoses, and the overall balance of benefits and risks was unfavourable. The oestrogen-only trial also stopped in 2004, but not because of more breast cancer; the increased risk of stroke played an important role.[1][12]
Which criticisms are justified?
- The average participant was older than many women starting treatment today. The findings do not automatically provide a personal risk profile for a healthy 51-year-old woman with severe hot flushes.
- Specific tablets were tested. Not all hormones, doses and routes of administration. Certainly not a direct comparison with today's combination of an estradiol patch and progesterone.
- Different health outcomes have too often been lumped together. Age and the timing of starting treatment matter to the overall assessment, particularly for cardiovascular disease. That does not prove that starting early removes the breast cancer risk.
There were also limitations, such as participants stopping their assigned treatment and less complete recording of diagnoses in later years. These limitations need to be taken into account, but they do not suddenly make a large randomised trial worthless. The later data still indicate an increased breast cancer risk with the combination studied.[2][6]
What did long-term follow-up show?
A 2020 publication in JAMA reported the results of long-term follow-up of the original groups. For breast cancer diagnoses, median follow-up was 18.9 years in the combined-therapy trial and 16.2 years in the oestrogen trial. The numbers below have been converted into average annual figures over that follow-up period.[2]
| Originally assigned treatment | Hormone group | Corresponding placebo group | Difference |
|---|---|---|---|
| CEE + MPA, uterus present | 45 | 36 | 9 more |
| CEE alone, after hysterectomy | 30 | 37 | 7 fewer |
Rounded from the reported annual percentages. The relative risk of a breast cancer diagnosis was approximately 28% higher with CEE + MPA and 22% lower with CEE alone; both differences were statistically significant. Each row has its own placebo group. They are not a clean test of the effect of adding MPA alone.
With combined therapy, breast cancer mortality did not differ statistically significantly from placebo. This does not mean that a difference in mortality was ruled out: the numbers were too small and the uncertainty intervals too wide for that. With CEE alone, both fewer diagnoses and fewer deaths from breast cancer were found.[2]
Long follow-up is not the same as long-term hormone use. The assigned treatment lasted a median of 5.6 years for CEE + MPA and 7.2 years for CEE alone. This was therefore not a trial of twenty years of uninterrupted treatment. Nor should you simply multiply the average annual differences by twenty.
Was hormone therapy discouraged unfairly, then?
Hormone use fell sharply after the first WHI results. The leap from “this specific combination has drawbacks” to “hormones are a bad choice for every woman” was too great. But the opposite narrative — “the research was wrong, so bioidentical hormones carry no risk” — is equally incorrect. Modern guidelines explicitly recognise hormone therapy as a treatment option; licensed bioidentical medicines are part of mainstream care.[3][6][10]
What about the recent US changes to warnings?
On 12 February 2026, the FDA confirmed that changes had been approved for six products, removing text about breast cancer, among other matters, from the most prominent warning panel, the boxed warning. This process began in November 2025. It is a change in US risk communication, not new research demonstrating zero risk. Nor does it mean that all warnings have disappeared from the full product information for every product.[13]
3. Are estradiol and progesterone safer, then?
Indications, not a guarantee
It is reasonable to argue that certain modern combinations have a more favourable profile than the older WHI combination. It has not been proved that this means they carry no additional breast cancer risk.
Where does the optimism come from?
The French E3N study followed 80,377 women. Oestrogen plus progesterone was associated with less breast cancer than oestrogen plus a group of other progestogens. Compared with women who had never used hormone therapy, the estimated relative risk was:
- 1.00 for oestrogen plus progesterone; uncertainty interval 0.83–1.22;
- 1.69 for oestrogen plus the other progestogens studied, excluding dydrogesterone; uncertainty interval 1.50–1.91.
A value of 1.00 means that the best estimate showed no difference. However, the uncertainty interval did not rule out, for example, an increase of around 22%. This is an average across the use studied, not a guarantee for every duration of use.[4]
A 2016 systematic review also found a lower risk with oestrogen plus progesterone than with oestrogen plus synthetic progestogens. However, it was based on only three observational studies. Moreover, “lower than with another treatment” is not the same as “as low as with no treatment”.[5]
Why is this not yet firm evidence?
An observational study compares existing users. Women are not randomly assigned to their treatments: they may differ in health, lifestyle, previous treatments and check-ups. Statistical adjustments help, but do not remove all differences. And large numbers of women in the overall study do not automatically mean there is extensive data on many years of use of one specific combination.
A large international analysis from 2019 found no convincing general exception for micronised progesterone with longer-term use. The US Menopause Society therefore states that some, but not all, observational data suggest a lower risk, and that randomised trials are needed to confirm this. A 2026 review article also identifies this lack of randomised evidence for differences between progestogens.[3][6][11]
The current NICE guideline is cautious: there is insufficient evidence to establish whether the additional breast cancer risk with combinations containing micronised progesterone or dydrogesterone differs from that with other progestogens. This does not mean that all treatments have been proved equally risky; it means that a difference is not yet sufficiently certain.[8]
Why is Dutch information sometimes more reassuring?
For estradiol plus progesterone, Thuisarts states that use for less than five years probably does not increase the risk of breast cancer. This Dutch patient information therefore interprets the findings more optimistically than NICE. The word probably, the specific combination and the limited duration of use are essential. This is not evidence for safe use indefinitely, nor a guarantee of a risk-free period up to exactly five years after starting treatment.[10]
The information on stopping also differs: Thuisarts mentions a return to the previous risk of disease after two years. For breast cancer after combined systemic therapy, that is not a reliable general assurance: NICE and the large 2019 analysis describe how some of the additional risk may persist for more than ten years. The duration and type of previous use matter.[3][8][10]
The fairest summary: there are reasons to consider licensed estradiol and micronised progesterone, but not to leave breast cancer out of the treatment discussion.
4. How large is the risk in straightforward numbers?
“A higher risk” can easily sound like “a high chance”. So always ask: how many additional cases, in how many women, over how many years, with which treatment? A relative increase of 25% would, for example, turn a risk of 4 in 1,000 into 5 in 1,000 — not 254 in 1,000. This is only an arithmetic example, not a personal estimate.
An estimate for women starting around age 50
The 2019 international analysis in The Lancet combined data from observational studies involving more than 100,000 women who developed breast cancer. The researchers produced the estimate below for women of average weight in Western countries. They assumed five years of hormone use starting at age 50, and counted breast cancer over the twenty years from age 50 to 69 inclusive.[3]
| Use starting at age 50 | Total | Additional cases compared with never using |
|---|---|---|
| No hormone therapy | 63 | — |
| 5 years of oestrogen alone | 68 | 5 |
| 5 years of oestrogen + a progestogen for part of each month | 77 | 14 |
| 5 years of oestrogen + a daily progestogen | 83 | 20 |
How to read these figures: these are rounded population estimates, assuming that the associations found are largely causal. They are not direct randomised comparisons, nor are they specific figures for an estradiol patch with micronised progesterone. The 63 per 1,000 is the baseline risk chosen for this calculation, not the lifetime risk in the Netherlands. In this analysis, ten years of use resulted in approximately twice as many additional cases as five years of use.
But did WHI not find less breast cancer with oestrogen alone?
Yes. This is where the research findings differ. WHI studied a particular medicine in women without a uterus, often at a different point after menopause. Observational studies involve other groups and treatments and may be affected by bias. These differences make the precise size of the effect uncertain. NICE summarises the evidence for oestrogen alone as showing little or no increase in breast cancer risk. This is not the same as proven general protection from every oestrogen.[2][8]
Do not add together figures from different studies. An annual WHI figure for women who were older on average and a twenty-year estimate starting at age 50 answer different questions. Giving an exact personal figure would not be justified without individual medical details — and without sufficient long-term research into the precise treatment.
5. A sound assessment is about more than breast cancer
Relief can be very valuable
Systemic hormone therapy — treatment that acts throughout the body — is highly effective against hot flushes and night sweats. If these disrupt sleep, sleep may also improve. Hormone therapy slows bone loss and reduces the risk of fractures during use. Severe symptoms and loss of quality of life are valid reasons in their own right to discuss treatment.[6]
The treatment goal matters
A favourable balance when treating symptoms does not mean that all women should take hormones. Guidelines do not recommend hormone therapy solely to prevent heart disease or dementia. “Restoring your hormones” is not sufficient evidence to support such claims.[8]
A patch or gel: a lower thrombosis risk does not mean no breast cancer risk
According to the available data, oestrogen delivered through the skin carries a lower risk of thrombosis than oestrogen tablets. Thrombosis is a blood clot that can block a blood vessel. NICE describes no increased risk of venous thrombosis with administration through the skin. The route of administration is also relevant to stroke. But for breast cancer, a patch or gel has not been shown to remove any additional risk. “Safer” should therefore always be followed by: for which outcome?[6][8]
If the uterus is present, its lining needs protection
Systemic oestrogen without sufficient progestogen can cause the lining of the uterus to keep growing and increase the risk of cancer of that lining, known as endometrial cancer. A suitable combination is therefore usually needed when the uterus is present. After complete removal of the uterus, oestrogen alone can usually be used; exceptions, for example because of a person's medical history, should be discussed with the doctor.[6][8]
An important practical risk: a custom-made progesterone cream applied to the skin has not been shown to provide sufficient protection for the uterine lining. Absorption may be too variable. The British Menopause Society warns about this. This is different from a licensed estradiol gel combined with an appropriately prescribed progestogen.[9]
Do not reduce or stop progesterone on your own to lower breast cancer risk. The dose, regimen and duration of protection must be appropriate for the oestrogen treatment.
A local vaginal treatment is a different category
Low-dose vaginal oestrogen for dryness or pain during sex is absorbed into the bloodstream only in small amounts and has a different risk profile from systemic hormones for hot flushes. Additional progestogen is usually not needed with this treatment. It does not treat hot flushes. If there is a history of breast cancer, local use also requires a separate assessment, involving the oncologist if necessary. Not every vaginal product is automatically low-dose or acts exclusively locally.[6][8]
Age, medical history and duration of use
According to the Menopause Society, the balance of benefits and risks is often favourable for troublesome symptoms when treatment starts before age 60 or within ten years of menopause, provided there are no medical reasons to advise against treatment. This is primarily a statement about the overall assessment: it does not remove the possible breast cancer risk. When starting at an older age or much later after menopause, cardiovascular risks, among others, weigh more heavily.[6]
A history of breast cancer, unexplained vaginal bleeding, previous thrombosis, stroke, heart attack or serious liver problems may be a reason not to use systemic hormones or to seek specialist advice first. A family history of breast cancer is not the same as having had breast cancer yourself, but it must be taken into account when assessing baseline risk.
There is no universal rule that everyone must stop after exactly five years, nor is there blanket approval for indefinite use. Discuss the lowest dose that provides sufficient relief, and regularly reassess whether the benefits outweigh the risks. NICE recommends a review after about three months and then annually. With very early menopause, especially before age 40, the balance is different; separate recommendations apply.[6][8]
6. Common claims: fact or oversimplification?
| The claim | Verdict | What can be said |
|---|---|---|
| “The older research was wrong, so the danger is made up.” | Incorrect | There was valid criticism of the broad application of the results, but the increased risk from the combination studied has not disappeared. |
| “WHI proves that every bioidentical regimen is dangerous.” | Also incorrect | The precise modern regimen was not tested. The findings cannot be transferred directly. |
| “Bioidentical means natural, so it is safe.” | Incorrect | The structure alone does not adequately tell us about risks at a particular dose, in a particular combination and over a particular duration of use. |
| “Progesterone seems to have a more favourable profile than some other progestogens.” | Plausible, still uncertain | There are favourable observational data, but no sufficient evidence of zero additional breast cancer risk. |
| “A patch gives you no additional breast cancer risk.” | Not demonstrated | The advantage of delivery through the skin mainly concerns thrombosis and possibly stroke, not a proven elimination of breast cancer risk. |
| “Hormone therapy can be a good choice for menopausal symptoms.” | Correct | For many women it can, after an individual assessment. A small risk may be acceptable in return for substantial symptom relief. |
| “A saliva test makes a hormone mixture precisely tailored and safe.” | Unsupported | Routine saliva or urine tests to determine doses of such mixtures are not supported by reliable evidence. A blood test may sometimes have a specific medical purpose, but it does not establish safety with regard to cancer.[6][7] |
7. What does this mean for you in practice?
The first step is not to choose between “the hormones are dangerous” and “they cannot do any harm”. The first step is to establish exactly what you are using and how much benefit you get from it. A photograph of the packaging or a medication list from the pharmacy gives the discussion a concrete starting point.
- Which active ingredients, dose and route of administration do you use? Estradiol, progesterone, another progestogen, a combination, or other hormones as well? Licensed or specially compounded? Local vaginal treatment only, or systemic?
- Do you still have a uterus? If so, is protection of the uterine lining appropriate for your oestrogen dose and regimen?
- When did you start, and how long have you been using the medicines? Note your age, the timing of your last natural menstrual period and any previous hormone treatments.
- How much do the hormones help? For example, fewer hot flushes, better sleep, being able to work again or less pain. The severity of the original symptoms must be explicitly included in the assessment.
- What is your personal baseline risk? Discuss any breast abnormalities or breast cancer you have had, family history, thrombosis, cardiovascular disease, blood pressure, smoking, alcohol and other relevant conditions.
- What do we know about this particular regimen, and what do we not know? Ask whether “safer” refers to breast cancer, thrombosis or endometrial cancer. Where possible, ask for absolute numbers and the corresponding time period.
- What is the follow-up plan? When should treatment be reviewed, which symptoms should be reported, and which alternatives would be appropriate if the balance changes? Participation in the routine national screening programme remains important; screening does not remove the risk associated with treatment.
Have a new breast lump, nipple change or unexplained/new vaginal bleeding assessed. This does not automatically mean cancer, but it should not be dismissed as “just the hormones”.
A balanced way to discuss this with your clinician:
“It is true that the older research was interpreted too broadly and that modern treatments may be different. But bioidentical is not a guarantee of zero risk. Let us look at whether the symptom relief from your specific treatment continues to outweigh the risks for you.”
8. How this review was put together
This is a focused exploration of the literature, not a full systematic review or an individual medical assessment. The emphasis is on original studies and major professional guidelines, not commercial websites or individual opinion pieces. Europe PMC was also searched specifically for publications from 2024–2026 on micronised progesterone and breast cancer. Sources were accessed on 20 September 2026; publication years are listed below.
- Randomised trials: random allocation makes treatment groups more comparable on average; strong for establishing cause and effect, but only for the treatments and population studied.
- Observational research: valuable for long-term use and different preparations, but more susceptible to differences between users and non-users.
- Guidelines: weigh the evidence as a whole. Where authoritative sources differ, this has been made explicit above rather than choosing a single reassuring or alarming source.
The NHG guideline itself could not be accessed directly during this review because of technical issues. However, the publicly available patient information from Thuisarts, supported by NHG and FMS, was read. The conclusions are also supported by international guidelines and research articles that were consulted directly.
Sources and what they do and do not show
- WHI — original combined-therapy trial (2002). Rossouw et al., Risks and benefits of estrogen plus progestin in healthy postmenopausal women. JAMA. Randomised comparison of CEE + MPA with placebo. Not a test of an estradiol patch plus micronised progesterone.
- WHI — long-term follow-up of both trials (2020). Chlebowski et al., Association of Menopausal Hormone Therapy With Breast Cancer Incidence and Mortality During Long-term Follow-up of the Women's Health Initiative Randomized Clinical Trials. JAMA. Makes the important distinction between CEE alone after hysterectomy and CEE + MPA when the uterus is present.
- International analysis of observational studies (2019). Collaborative Group on Hormonal Factors in Breast Cancer, Type and timing of menopausal hormone therapy and breast cancer risk. The Lancet. Supports the twenty-year estimates, the role of duration of use and the uncertainty about differences between preparations. Associations are not the same as effects demonstrated through randomisation.
- French E3N cohort (2008; online 2007). Fournier et al., Unequal risks for breast cancer associated with different hormone replacement therapies. Breast Cancer Research and Treatment. An important source for a possibly more favourable profile of progesterone and dydrogesterone; observational, so not conclusive evidence of an absence of risk.
- Systematic review of progesterone versus synthetic progestogens (2016). Asi et al., Progesterone vs. synthetic progestins and the risk of breast cancer: a systematic review and meta-analysis. Systematic Reviews. Only three observational studies; a comparison between treatments, not evidence that progesterone combinations are as safe as no treatment.
- The Menopause Society / NAMS — treatment guideline (2022). The 2022 hormone therapy position statement of The North American Menopause Society (PDF). Assessment of benefits and risks, age and timing of treatment initiation, contraindications, local therapy, bioidentical preparations and missing randomised data on breast cancer risk.
- ACOG — consensus on compounded preparations (2023). Compounded Bioidentical Menopausal Hormone Therapy. Clinical Consensus No. 6. No evidence supporting a routine preference for specially compounded mixtures over available licensed products. US regulatory context.
- NICE — Menopause: identification and management, NG23. Recommendations; particularly section 1.6, tables 1 and 2, and sections 1.5, 1.8 and 1.9. The breast cancer recommendations used here are from the 2024 revision; the current web version was consulted. Explicitly states that there is insufficient evidence to establish a difference in breast cancer risk between micronised progesterone or dydrogesterone and other progestogens.
- British Menopause Society — protection of the uterine lining (May 2026). Progestogens and endometrial protection (PDF). Protection must match the oestrogen dose and regimen; warns that absorption and protection provided by compounded progesterone creams are not sufficiently supported by evidence.
- Thuisarts — Dutch patient information (updated 17 August 2026). I am considering hormone therapy for hot flushes during menopause. Practical Dutch context. Differences from NICE regarding certainty and residual risk after stopping are explicitly addressed in this review.
- Recent review article (2026). Menopausal hormone treatment and breast cancer. The bibliographic details and abstract were consulted via Europe PMC (PMID 41619758), not the full paywalled text. Confirms that a lower risk with progesterone/dydrogesterone is based on observational data and has not been established through randomised comparative trials.
- WHI — original oestrogen-only trial (2004). Anderson et al., Effects of conjugated equine estrogen in postmenopausal women with hysterectomy. JAMA. The authors' abstract was consulted via Europe PMC (PMID 15082697). No demonstrated increase in breast cancer in this trial, but more strokes.
- FDA — confirmation of labelling changes (12 February 2026). FDA Approves Labeling Changes to Menopausal Hormone Therapy Products. Official confirmation for six products, not a new safety trial. The full revised product information for all individual products was not examined for this review.