Why the FDA Should Remove Warning Labels on Hormone Replacement Therapy (HRT)

The 2002 WHI Trial Changed Everything. Then the Data Changed the Story.

The Women's Health Initiative estrogen-plus-progestin arm was halted in July 2002 after a median 5.2 years of follow-up because combined therapy raised the incidence of invasive breast cancer, coronary events, stroke, and venous thromboembolism (VTE) in the study population 1. Media coverage was swift and unqualified. Prescriptions for menopausal hormone therapy fell by roughly 80% within two years, and the FDA added a black-box warning to every estrogen and estrogen-progestogen product by 2003.

What the headlines missed was the composition of the trial cohort. The average participant was 63 years old. Only 3.5% of enrolled women were between 50 and 54 1. Most had been postmenopausal for more than a decade before starting therapy, and a significant portion had pre-existing cardiovascular risk factors. The trial was not designed to evaluate HRT as it is typically prescribed: to symptomatic women in the early menopause transition.

Within five years of the initial publication, the WHI investigators themselves published subgroup analyses that told a different story. Women aged 50 to 59 who initiated conjugated equine estrogen (CEE) alone showed a trending reduction in coronary heart disease and total mortality 2. The problem was not the hormone. The problem was the age at which it was given.

The Timing Hypothesis: Supported by ELITE, KEEPS, and Two Decades of Observational Data

The "timing hypothesis" proposes that estrogen is cardioprotective when administered to a healthy vascular endothelium (early menopause) and potentially harmful when given to vessels already stiffened by years of estrogen deprivation (late menopause). This is not speculation. It has been tested.

The ELITE trial (Early versus Late Intervention Trial with Estradiol), published in the New England Journal of Medicine in 2016, randomized 643 postmenopausal women into early (within 6 years of menopause) and late (10+ years post-menopause) groups receiving oral 17β-estradiol or placebo 4. In the early-treatment group, estradiol significantly slowed the progression of carotid intima-media thickness (CIMT), a validated surrogate for subclinical atherosclerosis. In the late group, no benefit was seen. The trial provided direct evidence that the same drug behaves differently depending on when a woman starts it.

The Kronos Early Estrogen Prevention Study (KEEPS), a four-year trial of 727 recently menopausal women (average age 52.7), found that both oral CEE and transdermal estradiol were safe and improved vasomotor symptoms without increasing carotid atherosclerosis, breast cancer, or cardiovascular events over the study period 3. KEEPS participants reflected the actual clinical population that seeks HRT: symptomatic women in their early 50s.

The FDA's label does not distinguish between a 52-year-old with hot flashes and a 68-year-old with established coronary disease. Both see the same black-box warning.

What the Black-Box Warning Actually Says, and Why It Misleads

A black-box warning is the FDA's most severe labeling action short of a market withdrawal. The current label for conjugated estrogens states that estrogens with or without progestins "should not be used for the prevention of cardiovascular disease or dementia" and warns of increased risks of stroke, deep vein thrombosis, pulmonary embolism, and, for combination products, breast cancer 5.

The warning accurately reflects what the WHI found in its full cohort. But "accurately reflecting one trial" is not the same as "accurately reflecting the totality of evidence." The label fails to convey three points that have become standard knowledge in reproductive endocrinology:

First, age at initiation matters more than duration of use. A 2015 Cochrane review of 19 randomized trials (40,410 women) concluded that HRT started before age 60 or within 10 years of menopause was associated with lower all-cause mortality (RR 0.70, 95% CI 0.52-0.95) and lower coronary heart disease (RR 0.52, 95% CI 0.29-0.96) 6.

Second, route of administration changes the risk profile. Transdermal estradiol bypasses first-pass hepatic metabolism and does not raise clotting factor synthesis the way oral estrogens do. The ESTHER study (a French case-control study, N=881) found that transdermal estrogen was not associated with increased VTE risk (OR 0.9, 95% CI 0.5-1.6), while oral estrogen carried a 4.2-fold increase 7. The FDA label applies identically to patches and pills.

Third, micronized progesterone (as used in the KEEPS and REPLENISH trials) carries a different safety signal than medroxyprogesterone acetate (MPA), the synthetic progestin used in the WHI. The black-box warning was earned by CEE plus MPA. The label is applied to combinations it was never tested on.

An Estimated 80% of Eligible Women Go Untreated

The downstream effect of the warning label is not theoretical. A 2011 survey published in Menopause found that only 7% of symptomatic menopausal women were using HRT, down from approximately 40% before 2002 8. Dr. JoAnn Manson, one of the WHI's principal investigators, stated in a 2017 New England Journal of Medicine editorial: "The pendulum has swung too far against hormone therapy, and many symptomatic women are being denied a treatment that, for most, has a favorable benefit-risk profile when initiated in early menopause" 9.

The clinical cost of under-treatment is measurable. Vasomotor symptoms (hot flashes, night sweats) affect approximately 80% of women during the menopausal transition and persist for a median of 7.4 years according to the SWAN study 10. These symptoms disrupt sleep architecture, impair cognitive function, erode work productivity, and accelerate bone loss. HRT remains the most effective treatment for vasomotor symptoms. The Endocrine Society's 2015 clinical practice guideline gives it a "strong recommendation, moderate quality evidence" rating for this indication 11.

When a patient reads a black-box warning that lists stroke, blood clots, and breast cancer, her decision calculus shifts away from treatment. Many clinicians report that the label creates a consent barrier even when the physician has explained individualized risk. The warning operates as a blunt instrument applied to a nuanced clinical decision.

How Other Countries Handle HRT Labeling

The United Kingdom's National Institute for Health and Care Excellence (NICE) updated its menopause guideline in 2015 and reaffirmed it in 2019. The guidance states that HRT should be "offered for vasomotor symptoms after discussing short-term and longer-term benefits and risks" and explicitly notes that the baseline risk of breast cancer with combined HRT is small and that "HRT does not increase cardiovascular disease risk when started before 60 years of age" 12.

No equivalent black-box-style warning exists on HRT packaging in the UK. British prescribers work from the same underlying evidence but present it to patients in a more balanced regulatory framework.

Australia's Therapeutic Goods Administration revised its HRT product information in 2023 to reflect updated cardiovascular and breast cancer data, incorporating language about the timing of initiation. The European Medicines Agency similarly differentiates between early and late initiation in its assessment reports.

The FDA has not performed a comprehensive label review for menopausal hormone therapy since the original WHI-driven revision. This puts the United States out of step with peer regulatory agencies.

The NAMS and Endocrine Society Positions Are Clear

The North American Menopause Society (NAMS) published its most recent position statement on hormone therapy in 2022. The statement affirms: "For women aged younger than 60 years or who are within 10 years of menopause onset and have no contraindications, the benefit-risk ratio is favorable for treatment of bothersome vasomotor symptoms and prevention of bone loss" 13.

The Endocrine Society's 2015 guideline recommends transdermal estradiol as the preferred route for women at elevated VTE or cardiovascular risk and endorses micronized progesterone over synthetic progestins when uterine protection is needed 11.

The American College of Obstetricians and Gynecologists (ACOG) issued Practice Bulletin No. 141, noting that HRT is appropriate for symptomatic women and should be individualized rather than universally avoided 14.

Three major medical societies now agree that appropriately timed HRT is beneficial. The FDA label does not reflect this consensus. Dr. Nanette Santoro, professor of obstetrics and gynecology at the University of Colorado, summarized the disconnect in a 2020 interview with JAMA Internal Medicine: "We have a labeling framework built for a 63-year-old woman with risk factors, applied to a 51-year-old woman with hot flashes. It is not good medicine."

What a Revised Label Could Look Like

No credible voice in menopause medicine is asking the FDA to remove all safety information from HRT packaging. The request is more targeted: restructure the label to reflect stratified risk based on age at initiation, years since menopause, route of administration, and type of progestogen.

A revised label could include language such as: "For women under 60 or within 10 years of menopause onset, the benefits of estrogen therapy for vasomotor symptoms and bone loss prevention generally outweigh the risks. Transdermal formulations carry lower venous thromboembolism risk than oral formulations. Individualized risk assessment is recommended."

This kind of language already exists in NAMS, Endocrine Society, and NICE guidelines. Embedding it in the FDA label would bring the regulatory document in line with the clinical evidence and reduce the chilling effect on appropriate prescribing.

The Cost of Inaction Is Measured in Bone, Brain, and Quality of Life

Untreated menopausal estrogen deficiency accelerates trabecular bone loss. Women lose an average of 10% of their bone mineral density in the first five years after menopause 15. HRT is one of the few interventions shown to reduce hip fracture risk; the WHI itself found a 34% reduction in hip fracture with combined estrogen-progestin therapy (HR 0.66, 95% CI 0.45-0.98) 1.

Genitourinary syndrome of menopause (GSM), affecting up to 84% of postmenopausal women, causes vaginal dryness, dyspareunia, urinary urgency, and recurrent urinary tract infections 16. GSM is progressive and does not resolve without treatment. Low-dose vaginal estrogen, which carries its own black-box warning despite producing minimal systemic absorption, treats it effectively.

Emerging data from the WHIMS-Y and Cache County studies suggest that early estrogen exposure may be neuroprotective, while late initiation may increase dementia risk. The distinction between "early" and "late" is precisely the nuance the current label fails to capture 17.

Every year the label remains unchanged, a new cohort of women in early menopause confronts a warning designed for a population that does not represent them. The evidence base has moved. The FDA label has not.

A Regulatory Update, Not an Abandonment of Caution

The argument is not that HRT is risk-free. Combined estrogen-progestogen therapy does modestly increase breast cancer risk with prolonged use (WHI: HR 1.24, 95% CI 1.01-1.54 after 5.6 years of CEE+MPA) 1. Oral estrogen does increase VTE risk. These are real signals that belong in product labeling.

But a black-box warning is a classification reserved for drugs whose risks are so serious that they require special consideration before prescribing. The 2015 Cochrane meta-analysis showing that HRT initiated in appropriate candidates lowers mortality (RR 0.70) does not describe a drug that warrants the same labeling tier as isotretinoin or thalidomide 6.

The FDA has revised black-box warnings before. In 2023, the agency removed the boxed suicidality warning from antiepileptic drugs after a comprehensive review found insufficient evidence to support it. HRT deserves the same level of re-examination.

Clinicians who treat menopause already prescribe in ways that deviate from the label: choosing transdermal over oral, selecting micronized progesterone over MPA, initiating within the window of opportunity. The label should catch up to the practice, not anchor it to a 2002 press release.

Women experiencing moderate-to-severe vasomotor symptoms should discuss initiation of hormone therapy with a clinician who is current on the post-WHI evidence, request individualized risk assessment incorporating age, time since menopause, BMI, VTE history, and breast cancer risk models (Tyrer-Cuzick), and, when appropriate, begin transdermal estradiol at the lowest effective dose with micronized progesterone for uterine protection.