WHI E-alone Extension Data and What Happened After the Trial Ended

At a glance

Parameter	Detail
Trial	Women's Health Initiative Estrogen-Alone (WHI E-alone)
N	10,739 post-hysterectomy women aged 50-79
Intervention	Conjugated equine estrogen (CEE) 0.625 mg/day
Comparator	Matching placebo
Active phase duration	Median 7.2 years (stopped early February 2004)
Post-stopping follow-up	Extended to median 13.2 years cumulative
Primary endpoint	Coronary heart disease (nonfatal MI + CHD death)
Key result	No lasting CHD benefit; persistent breast cancer risk reduction (HR 0.79, 95% CI 0.65-0.97) through extended follow-up

Why the Extension Matters More Than the Original Trial

The original WHI E-alone trial was stopped early in 2004 after a median 7.2 years because CEE increased stroke risk without reducing CHD. That early stop left critical questions unanswered. Would the breast cancer reduction hold? Would cardiovascular signals shift with longer observation? Would mortality differences emerge?

The NIH funded a post-intervention follow-up phase. Participants who consented (7,645 of the original 10,739) were observed without study drug. Combined with the active phase, this produced cumulative data spanning over 13 years, enough time for latent effects to surface.

The Post-Stopping Follow-Up Design

After intervention ended, 78% of surviving participants enrolled in the WHI Extension Study (Extension I, 2005-2010) and subsequently Extension II (2010-2015). Outcome ascertainment used the same adjudication committees as the main trial. Participants were unblinded in 2004, which introduced potential detection bias for self-reported outcomes, though hard endpoints (death, hospitalized MI, cancer registry-confirmed diagnoses) remained strong.

The key post-intervention analysis published in JAMA in 2011 reported outcomes through a median cumulative follow-up of 11.8 years. A subsequent 2017 JAMA analysis extended this to 18 years of cumulative follow-up across both the active and post-stopping phases.

HealthRX.com Signal Durability Framework: What Held, What Faded, What Flipped

Tracking how trial signals evolve after drug discontinuation reveals which effects reflect lasting biological change and which depend on continuous exposure. For the WHI E-alone extension, we categorize each major outcome.

Signals That Held: Breast Cancer Reduction

The most striking finding from the original trial was a 23% reduction in invasive breast cancer (HR 0.77, 95% CI 0.59-1.01) that narrowly missed statistical significance. Through the post-intervention period, this reduction persisted and strengthened statistically.

This persistence is biologically notable. CEE appeared to reduce breast cancer incidence during active use, and the reduction did not rebound after stopping. The mechanism remains debated. One hypothesis involves estrogen-induced apoptosis of occult estrogen-sensitive tumor cells, a paradox explored in preclinical models where prolonged estrogen deprivation sensitizes certain breast cancer cells to estrogen-triggered cell death.

This stood in direct contrast to the WHI combined estrogen-plus-progestin arm, where breast cancer risk was elevated. The divergence confirmed that progestin, not estrogen, was the primary driver of breast cancer risk in combined HRT regimens. The Premarin FDA label now reflects this distinction in its clinical trial discussion sections.

Signals That Faded: Cardiovascular Outcomes

During the active phase, the original WHI E-alone analysis reported a nominally favorable CHD signal in women aged 50-59 (HR 0.63, 95% CI 0.36-1.08). This generated optimism that early-initiation estrogen might be cardioprotective, consistent with the "timing hypothesis."

Post-stopping data dampened that optimism.

The CHD benefit in younger women did not survive the post-intervention period. Once CEE was discontinued, event rates converged. This pattern is consistent with a pharmacologic (hemodynamic, lipid-mediated) effect that depends on continuous exposure rather than a lasting structural change in coronary vasculature.

The stroke excess seen during active treatment (overall HR 1.39, 95% CI 1.10-1.77 in the original trial) also resolved after stopping, returning to baseline rates within 2-3 years. This confirmed the stroke risk was exposure-dependent and reversible.

Signals That Stayed Null: All-Cause Mortality

All-cause mortality showed no significant difference during the active phase or through extended follow-up.

For women aged 50-59, a non-significant mortality reduction during the active phase (HR 0.73, 95% CI 0.47-1.13) attenuated further with extended follow-up. The 2017 cumulative analysis through 18 years confirmed no mortality difference in any age subgroup.

What the Original Trial Could Not Show

Hip Fracture Protection Was Real but Temporary

CEE reduced hip fractures during active use (HR 0.61, 95% CI 0.41-0.91). After stopping, fracture rates returned to placebo levels within 3 years. Bone mineral density gains accrued during treatment were lost at roughly 2-3% per year post-discontinuation, consistent with the known reversibility of estrogen's antiresorptive effects on bone.

This has direct clinical implications: CEE is not a "treat and stop" osteoporosis strategy. Women who discontinue need transition to bisphosphonates or other antiresorptive agents, a point now reflected in the Endocrine Society's 2019 guidelines on postmenopausal osteoporosis.

Cognitive Outcomes Remained Ambiguous

The WHI Memory Study (WHIMS) substudy found increased dementia risk in women over 65 receiving CEE alone. Post-stopping cognitive assessments showed no lasting impairment, but the substudy was underpowered to detect subtle effects. The age of the WHIMS cohort (65-79 at enrollment) limits extrapolation to women starting estrogen in their 50s, and no extension data specifically addressed this younger group with adequate power.

Diabetes Signal Emerged

A secondary analysis of the intervention phase found CEE reduced incident diabetes by 12% (HR 0.88, 95% CI 0.77-1.01). Post-stopping, this effect disappeared. Like the cardiovascular signal, the metabolic benefit required continuous exposure. Fasting glucose and insulin resistance markers returned to placebo levels within 1-2 years of discontinuation.

Limitations the Extension Data Inherited

Differential dropout. Only 78% of surviving participants consented to follow-up. Women who dropped out had higher baseline BMI, more smoking, and lower education. This introduces healthy-survivor bias favoring the follow-up cohort.

Unblinding. After 2004, participants knew their assignment. This could alter health behaviors (former placebo recipients starting commercial HRT, former CEE recipients continuing or stopping). Post-stopping HRT use was reported at roughly 5-6% in both groups, minimizing but not eliminating this concern.

Single dose, single formulation. The trial tested only CEE 0.625 mg/day. Whether transdermal estradiol, lower-dose oral estradiol, or other formulations would show similar long-term patterns is unknown. The 2017 NICE menopause guideline notes that transdermal estradiol may carry lower thrombotic risk than oral CEE, but no RCT of comparable size has tested this.

Age distribution. Mean enrollment age was 63.6 years. The 50-59 subgroup (n=1 to 637 in CEE arm) was too small for definitive conclusions on most endpoints. The promising signals in younger women generated hypotheses that remain unconfirmed by dedicated RCTs.

Clinical Translation: What Changed in Practice

The extension data shifted the conversation from "is estrogen safe" to "for whom, when, and how long." Three practice-level changes emerged:

First, the breast cancer finding removed a major objection to estrogen-alone therapy in hysterectomized women. Clinicians became more comfortable prescribing CEE (or estradiol) to symptomatic post-hysterectomy patients without adding a progestin. Current NAMS 2022 position statements cite the WHI E-alone extension data when noting that estrogen alone does not increase breast cancer risk.

Second, the fading of cardiovascular benefit after stopping undermined the concept of estrogen as a long-term cardioprotective strategy. The timing hypothesis survived as a framework for explaining age-dependent risk-benefit ratios, but the extension data showed that even well-timed estrogen did not leave a lasting cardiovascular imprint.

Third, the reversibility of bone protection reinforced the need for sequential therapy planning. Starting estrogen for menopausal symptoms, then transitioning to a bisphosphonate for skeletal protection, became a common clinical sequence.

Frequently asked questions

›Did the WHI E-alone extension show estrogen prevents heart disease?

No. The modest CHD benefit observed in women aged 50-59 during active treatment faded after CEE was stopped. Cumulative follow-up through 13+ years showed no lasting cardiovascular advantage in any age group.

›How long did the breast cancer reduction last after stopping estrogen?

The reduction in invasive breast cancer persisted through at least 13 years of cumulative follow-up (HR 0.79, 95% CI 0.65-0.97), with the effect still present at 18-year analysis. The mechanism may involve irreversible effects on occult tumor biology.

›Was the stroke risk permanent after taking estrogen?

No. The elevated stroke risk during active CEE use resolved within 2-3 years of stopping. The effect was exposure-dependent and fully reversible.

›Why was the WHI E-alone trial stopped early?

The Data Safety Monitoring Board stopped the trial in February 2004 because CEE increased stroke risk (HR 1.39) without reducing the primary endpoint of CHD. The global index did not favor continuation.

›Does the WHI extension support starting estrogen in your 50s?

The age-stratified data suggest a more favorable risk-benefit profile for women aged 50-59, but the subgroup was small and the cardiovascular benefit did not persist after stopping. Current guidelines support use for symptomatic women near menopause, not as a preventive cardiovascular strategy.

›Did bone density improvements last after stopping estrogen?

No. Bone mineral density gains were lost within approximately 3 years of discontinuation, and hip fracture rates returned to baseline. Transition to another antiresorptive agent is recommended.

›How does WHI E-alone differ from the WHI combined HRT trial?

The combined trial (CEE + medroxyprogesterone acetate) showed increased breast cancer risk, while the estrogen-alone arm showed decreased risk. This confirmed progestin as the driver of HRT-associated breast cancer. Cardiovascular risk patterns also differed between arms.

›What percentage of WHI participants continued into the extension study?

Approximately 78% of surviving participants (7,645 of the original 10,739) consented to the post-intervention follow-up phase.

›Does the WHI extension data apply to transdermal estradiol?

Not directly. The trial used only oral conjugated equine estrogen at 0.625 mg/day. Transdermal estradiol has different pharmacokinetics and may carry lower thrombotic risk, but no comparable RCT exists for long-term outcomes.

›Did estrogen reduce diabetes risk long-term?

During active use, CEE reduced incident diabetes by approximately 12%. This benefit disappeared after stopping, indicating the metabolic effect required continuous exposure.

References

1. Anderson GL, Limacher M, Assaf AR, et al. Effects of conjugated equine estrogen in postmenopausal women with hysterectomy: the Women's Health Initiative randomized controlled trial. JAMA. 2004;291(14):1701-1712. PubMed
2. LaCroix AZ, Chlebowski RT, Manson JE, et al. Health outcomes after stopping conjugated equine estrogens among postmenopausal women with prior hysterectomy: a randomized controlled trial. JAMA. 2011;305(13):1305-1314. PubMed
3. Manson JE, Aragaki AK, Rossouw JE, et al. Menopausal hormone therapy and long-term all-cause and cause-specific mortality: the Women's Health Initiative randomized trials. JAMA. 2017;318(10):927-938. PubMed
4. The 2022 Hormone Therapy Position Statement of The North American Menopause Society. Menopause. 2022;29(7):767-794. PubMed
5. Premarin (conjugated estrogens) prescribing information. Pfizer. FDA Label
6. Eastell R, Rosen CJ, Black DM, et al. Pharmacological management of osteoporosis in postmenopausal women: an Endocrine Society clinical practice guideline. J Clin Endocrinol Metab. 2019;104(5):1595-1622. PubMed