Medroxyprogesterone (Provera) in Women's HRT: Dosing, Risks, and How It Compares to Other Progestogens

Medroxyprogesterone acetate (MPA) is a synthetic progestogen, marketed for this use as Provera (oral tablets), and it is also the progestogen in the combined product Prempro (conjugated equine estrogens plus MPA). It is a different molecule from Depo-Provera, the injectable contraceptive form of the same drug used at much higher doses for a different purpose, and it is chemically distinct from micronized progesterone (Prometrium), which is structurally identical to the progesterone the ovary makes. In women's HRT, MPA's job is narrow: taken alongside estrogen in a woman who still has a uterus, it prevents estrogen-driven overgrowth of the uterine lining. That endometrial-protection effect is well established. What is genuinely contested, and what should change a reader's decision, is whether MPA specifically carries a different breast cancer signal than micronized progesterone when both are paired with estrogen for years at a time.

The core answer: in women with a uterus, adding a progestogen to estrogen therapy is medically necessary to prevent endometrial hyperplasia and cancer, and MPA at FDA-labeled doses (2.5 mg daily continuous, or 5 to 10 mg for 12 to 14 days per cycle) does this effectively. Separately, the Women's Health Initiative (WHI) trial found that oral conjugated estrogen plus MPA was associated with an increase in invasive breast cancer risk after roughly five years of use, while the estrogen-only arm (no MPA, used only in women without a uterus) did not show the same signal. Observational cohort data, most notably the French E3N study, has reported a smaller or absent breast cancer association when micronized progesterone is used instead of a synthetic progestogen. No large head-to-head randomized trial has directly compared MPA against micronized progesterone for breast cancer outcomes, so this remains an inference from combining a randomized trial (MPA vs. placebo) with observational cohort data (MPA vs. micronized progesterone), not a single controlled comparison.

What MPA is approved to do, and what it is not

MPA's FDA-approved role in hormone therapy is endometrial protection: preventing the endometrial hyperplasia that unopposed estrogen can cause in a woman who has not had a hysterectomy. It is also FDA-approved, at different doses, for secondary amenorrhea and abnormal uterine bleeding, and at a much higher injectable dose (Depo-Provera) for contraception, a distinct use case with different pharmacokinetics that should not be confused with oral HRT dosing. This article covers oral MPA as used in combination HRT, not the injectable contraceptive product.

Using estrogen without any progestogen in a woman with an intact uterus is not a minor omission. Unopposed estrogen is a recognized cause of endometrial hyperplasia and, with prolonged use, endometrial cancer; this is precisely why every major guideline requires progestogen co-administration in women with a uterus, and it is the reason MPA (or an alternative progestogen) is prescribed at all.

Approved dosing regimens

MPA for HRT is used in two regimens, and the right one depends mainly on time since the last period.

Continuous-combined: MPA 2.5 mg daily alongside daily estrogen. This is generally reserved for women at least 12 months past their final menstrual period, because it is intended to produce an atrophic, non-bleeding endometrium over time rather than a scheduled bleed. Irregular spotting is common in the first three to six months.

Sequential (cyclic): MPA 5 to 10 mg for 12 to 14 consecutive days each month, with estrogen continued daily. This produces a scheduled withdrawal bleed. Older endometrial-histology studies from the 1980s established that shorter progestogen courses (around 10 days) leave a meaningful minority of women with residual endometrial proliferation, which is why 12 to 14 days became the standard; readers considering a shorter or lower-dose regimen should confirm the specific evidence with their prescriber rather than assume any duration is adequate.

FDA-approved labeling for Provera specifies these dose ranges for endometrial protection when used with estrogen. Anyone checking exact current label language should consult the FDA's own Provera and Prempro labeling pages directly, since prescribing information can be revised and the details here should not substitute for the current label.

What the WHI trial actually showed, and its limits

The WHI combined-hormone trial randomized postmenopausal women with a uterus to oral conjugated equine estrogen 0.625 mg plus MPA 2.5 mg (the Prempro regimen) or placebo. After a mean follow-up of about 5.6 years, the trial reported an increase in invasive breast cancer in the combined-hormone group. A separate WHI arm enrolled women without a uterus (so no progestogen was needed) and randomized them to estrogen alone or placebo; that arm did not show the same breast cancer increase over roughly seven years of follow-up. This estrogen-only versus combined-hormone contrast is the main reason clinical attention shifted toward the progestogen component as a possible driver of the signal, rather than estrogen itself.

Three limits matter for anyone applying this to a current decision:

• The trial tested one specific combination: oral conjugated equine estrogen plus oral MPA. It did not test transdermal estradiol, other estrogen doses, micronized progesterone, or shorter treatment durations.
• WHI participants were, on average, in their early-to-mid sixties and roughly a decade past menopause at enrollment, older, on average, than the woman in her early-to-mid fifties who typically starts HRT for menopausal symptoms today. Whether the same magnitude of risk applies to someone starting near the time of menopause is a genuinely open question (the "timing hypothesis," discussed below), not a settled fact.
• WHI compared MPA-containing HRT to no HRT. It does not, by itself, tell you how MPA compares to micronized progesterone; that comparison relies on separate observational data.

Readers who want the exact hazard ratios and confidence intervals should look up the original WHI publications (JAMA, 2002 and 2004) directly rather than relying on secondhand figures, since precise numbers are easy to misquote and matter for shared decision-making.

MPA versus micronized progesterone: how strong is the difference, really?

Micronized progesterone (Prometrium, 100 mg and 200 mg capsules) has become many prescribers' default progestogen, largely on the strength of observational cohort data suggesting a more favorable breast cancer association than synthetic progestogens including MPA. The most frequently cited source is the French E3N cohort, which followed a large group of postmenopausal women and reported a materially higher breast cancer association with synthetic progestogens (including MPA) than with micronized progesterone when each was combined with estrogen. This is observational, not randomized, evidence: it cannot fully exclude the possibility that women prescribed micronized progesterone differed in other ways (prescribing era, adherence, baseline risk) from women prescribed MPA. It is a real and widely cited signal, but it sits below trial-level evidence in the hierarchy, and no randomized trial has been designed or powered to settle the question directly.

A separate randomized trial (the Kronos Early Estrogen Prevention Study, or KEEPS) compared oral and transdermal estrogen, each paired with oral micronized progesterone, against placebo in recently menopausal women, primarily to study cardiovascular surrogate outcomes rather than breast cancer. It is relevant to the estrogen-route discussion below but does not itself compare MPA to progesterone.

A 2025 meta-analysis of randomized controlled trials examined MPA's effect on apolipoprotein and lipoprotein(a) concentrations in postmenopausal women (PubMed 39542236). This addresses a lipid and cardiovascular-marker question, not breast cancer, and it is worth reading directly before making any claim about MPA's cardiovascular lipid profile, since the clinical significance of apolipoprotein and Lp(a) shifts for hard cardiovascular outcomes is not established by this kind of surrogate-marker analysis alone.

There are legitimate, non-breast-cancer reasons to still choose MPA over micronized progesterone:

• Micronized progesterone capsules are formulated with peanut oil and are contraindicated in a peanut allergy.
• MPA is available as a low-cost generic, which matters when coverage or cost is the deciding factor.
• Micronized progesterone has a sedative effect for some women, related to its activity at GABA-A receptors, which some find intolerable; MPA does not share this effect to the same degree.
• MPA has a longer track record in some sequential regimens where a specific bleeding pattern is desired.

A decision framework, not a substitute for individualized prescribing

The choice of progestogen and estrogen route is rarely a single fact but a set of tradeoffs. This framework lays out the recurring decision points; it is a starting point for a conversation with a prescriber, not a protocol to self-apply.

Estrogen delivery choices that pair with either progestogen

The progestogen is one half of combination HRT. Estrogen delivery route matters independently, mainly because of what happens to estrogen after absorption.

Oral estradiol is the lowest-cost and most familiar option. Because it passes through the liver before reaching systemic circulation, it increases hepatic production of clotting factors and sex hormone-binding globulin. Multiple observational studies have associated oral estrogen, but not transdermal estrogen, with an increased risk of venous thromboembolism. Women with a personal or family history of blood clots, known thrombophilia, or significant obesity are generally steered toward a transdermal route for this reason; this is a route-selection issue, not something MPA or micronized progesterone changes.

Transdermal estradiol (patch) bypasses first-pass liver metabolism and, in large observational cohorts, has not shown the same VTE association as oral estrogen. Patches are typically changed once or twice weekly depending on the brand; local skin irritation is a common but usually manageable side effect.

Estradiol gel (Divigel and similar products) is applied daily to the skin, avoids hepatic first-pass effects similarly to the patch, and is FDA-approved for moderate to severe vasomotor symptoms. Skin-to-skin transfer to a partner or child is a real concern until the gel has fully dried, and the application site should be covered once dry.

Estradiol spray (Evamist) delivers estradiol via a metered spray to the forearm, is titrated from one spray up to three per day based on symptom control, and carries the same accidental-transfer caution as the gel until it has dried.

Any of these estrogen routes can be paired with either MPA or micronized progesterone. The route decision (oral versus transdermal) is driven mainly by clotting risk and personal tolerance; the progestogen decision (MPA versus micronized progesterone) is driven mainly by allergy status, cost, sedation tolerance, and the reader's own weighting of the observational breast cancer data discussed above.

Side effects

MPA's most commonly reported side effects are breast tenderness, bloating, mood changes, acne, headache, and breakthrough bleeding, especially in the first one to three months of a continuous-combined regimen. MPA has mild androgenic and glucocorticoid activity that micronized progesterone largely lacks, which may explain why acne and mood-related complaints are reported somewhat more often with MPA in patient surveys. Some smaller studies have suggested better subjective sleep and mood outcomes with micronized progesterone compared with MPA-based regimens, but a specific trial claim from an earlier version of this article could not be verified against its cited source and has been removed rather than repeated; anyone weighing this tradeoff heavily should ask their prescriber for current comparative data rather than relying on this article's summary alone.

Persistent bleeding beyond six months on a continuous-combined regimen, or new bleeding after 12 months of amenorrhea, is not something to wait out. It warrants evaluation, typically transvaginal ultrasound with possible endometrial biopsy, to rule out hyperplasia or a polyp.

Who should not take MPA or combined HRT

The following are widely recognized contraindications to estrogen-progestogen HRT, consistent with FDA labeling for these products. This list is a starting point for a clinical conversation, not a substitute for a prescriber's full assessment of a woman's history:

• Known, suspected, or prior history of breast cancer
• Known or suspected estrogen- or progestogen-dependent malignancy
• Active or prior arterial thromboembolic disease (stroke, myocardial infarction)
• Active deep vein thrombosis or pulmonary embolism, or a prior history without ongoing anticoagulation
• Active liver disease with abnormal liver function
• Undiagnosed abnormal genital bleeding
• Known or suspected pregnancy

Anyone with these conditions, or uncertainty about whether one applies, should not start HRT without direct evaluation by a clinician. This article cannot substitute for that assessment.

Monitoring while on therapy

A baseline clinical breast exam and mammogram before or shortly after starting MPA-containing HRT, with continued annual mammography, is standard practice. Routine endometrial ultrasound is not required for a woman doing well on an adequate progestogen dose, but it becomes appropriate if breakthrough bleeding persists past six months or new bleeding appears after a year of amenorrhea. Blood pressure should be checked a few months after starting and then annually, since oral estrogen can raise blood pressure in some women through liver-mediated effects that transdermal routes largely avoid. A baseline lipid panel is reasonable before starting oral estrogen given its effect on triglycerides; this is less critical with transdermal routes in women without pre-existing high triglycerides.

Duration of use and the timing hypothesis

WHI enrolled women who were, on average, over a decade past menopause. The "timing hypothesis" holds that starting HRT closer to menopause onset, generally within about ten years or before age 60, carries a more favorable balance of benefits and risks than starting later. Post hoc, age-stratified analyses of WHI data are consistent with this pattern, but a post hoc subgroup analysis is weaker evidence than a trial specifically designed and powered to test the hypothesis, and it has not been confirmed by a dedicated randomized trial in that younger population. Treat the timing hypothesis as plausible and guideline-endorsed, not proven.

Major bodies, including the North American Menopause Society and the UK's National Institute for Health and Care Excellence (NICE), have taken the general position that for most women with bothersome menopausal symptoms, and no contraindication, the benefits of HRT outweigh the risks, while also noting that any breast cancer risk increase depends on duration of use (NICE guideline NG23). There is no universal maximum duration that applies to every woman. An annual reassessment of symptoms, quality of life, and updated personal risk factors, rather than a fixed stop date, is the approach most guidelines recommend.

What is established, what is plausible, and what is not established

Established: unopposed estrogen increases endometrial cancer risk in women with a uterus; adding an adequate dose of a progestogen, including MPA, largely eliminates that excess risk. Oral estrogen carries a higher VTE association than transdermal estrogen in observational data. The WHI trial found an increase in breast cancer with oral conjugated estrogen plus MPA after several years of use, and its estrogen-only arm did not show the same signal.

Plausible but unproven: that micronized progesterone carries a meaningfully lower breast cancer risk than MPA when both are used long-term with estrogen. This rests on observational cohort comparisons, not a randomized head-to-head trial, and residual confounding cannot be excluded. The timing hypothesis, that starting HRT near menopause onset is safer than starting a decade or more later, is biologically plausible and supported by subgroup analyses, but has not been confirmed in a purpose-built trial.

Not established: a universal duration limit for MPA-containing HRT that applies to all women; a precise, individualized breast cancer risk number that a specific reader can expect, since population-level hazard ratios do not translate directly into personal risk without knowing baseline risk, family history, and other factors.

Common questions

Is Provera the same as progesterone? No. Provera is medroxyprogesterone acetate, a synthetic molecule that activates progesterone receptors but is structurally different from the progesterone the body produces. Micronized progesterone (Prometrium) is body-identical progesterone in capsule form and has a different side-effect and, per observational data, possibly different long-term risk profile.

Can MPA be paired with a transdermal estrogen patch instead of an oral pill? Yes. MPA can be combined with any systemic estrogen delivery route, including a patch, gel, or spray. The estrogen route decision (oral versus transdermal) is largely separate from the progestogen decision, and transdermal routes are generally preferred for women with clotting risk factors regardless of which progestogen is used.

Does MPA cause weight gain? Weight change is commonly attributed to hormone therapy, but the WHI trial did not find a clear difference in weight gain between combined HRT and placebo groups. Much of the weight and body-composition change women notice around menopause reflects broader metabolic and activity changes rather than a direct drug effect, though this is difficult to fully disentangle for any individual.

How long should someone stay on MPA-containing HRT? There is no fixed limit that applies to everyone. Because the WHI breast cancer signal emerged after roughly five years of continuous oral estrogen plus MPA, many clinicians treat that point as a natural moment to revisit the risk-benefit balance, but continuing or stopping should reflect an individual's symptoms, risk factors, and preferences, reassessed at least annually.

What should happen if breakthrough bleeding continues past six months? It should be evaluated, not assumed to be normal adjustment. Transvaginal ultrasound, and often an endometrial biopsy, is the standard next step to rule out hyperplasia or a polyp before continuing the same regimen unchanged.

References

• U.S. Food and Drug Administration. Provera and Prempro prescribing information (consult current FDA label directly for exact current dosing and warnings): https://www.accessdata.fda.gov
• North American Menopause Society. 2022 Hormone Therapy Position Statement (consult the Menopause Society directly for exact current position statement text).
• National Institute for Health and Care Excellence. Menopause: diagnosis and management, NICE guideline NG23: https://www.nice.org.uk/guidance/ng23
• Women's Health Initiative combined-hormone and estrogen-alone trial results, originally published in JAMA (2002, 2004); consult JAMA's archive directly for exact figures and confidence intervals.
• E3N cohort study on hormone therapy type and breast cancer risk, published in Breast Cancer Research and Treatment; consult the original publication directly for exact risk estimates.
• Effect of medroxyprogesterone acetate on apolipoproteins and lipoprotein(a) concentrations in postmenopausal women: a meta-analysis of randomized controlled trials (2025): https://pubmed.ncbi.nlm.nih.gov/39542236/

This article summarizes publicly reported clinical trial and guideline findings for general education. It is not individualized medical advice and does not replace an evaluation by a qualified clinician, who can review a specific person's history, contraindications, and current FDA labeling before recommending a dose or regimen. Several precise figures and quotations from an earlier version of this page could not be verified against a primary source and have been removed or converted to general statements; readers or reviewers who need exact statistics should consult the primary publications listed above directly.