Was the ELITE trial randomized and blinded?

Yes. ELITE was a randomized, double-blind, placebo-controlled trial. Participants and investigators were masked to treatment assignment. Randomization was stratified by time since menopause (<6 years vs ≥10 years).

Why did ELITE use carotid IMT instead of heart attack rates?

Measuring actual cardiovascular events would require tens of thousands of participants followed for 10+ years. CIMT allowed a smaller trial to test the timing hypothesis with a validated imaging biomarker. The tradeoff is that surrogate findings do not guarantee clinical event reduction.

How does ELITE differ from the WHI?

WHI enrolled primarily older women (average age 63) and measured hard clinical outcomes over longer follow-up. ELITE enrolled both early and late postmenopausal women and used an imaging surrogate. The two trials address different questions at different timescales.

Did the ELITE trial test transdermal estrogen?

No. ELITE tested only oral 17β-estradiol at 1 mg daily. Results cannot be directly extrapolated to transdermal patches, gels, or sprays, which have different pharmacokinetic and hepatic profiles.

What was the dropout rate in ELITE?

Approximately 16% of participants discontinued the study drug before the end of the trial. The primary analysis used intention-to-treat methods, which included all randomized participants regardless of adherence.

Did ELITE show any benefit in late-postmenopausal women?

No. Women who were 10 or more years past menopause showed no significant difference in CIMT progression between estradiol and placebo groups. The treatment-by-timing interaction was statistically significant.

Has any trial replicated the ELITE finding?

No independent trial has replicated ELITE's primary result using the same design and population. KEEPS, which studied a similar age group, found no significant CIMT benefit with HRT, though it used different estrogen formulations and doses.

Should doctors prescribe HRT for heart protection based on ELITE?

Current guidelines do not recommend initiating HRT solely for cardiovascular prevention. ELITE supports the safety of early HRT from a vascular standpoint but does not constitute proof of clinical cardiovascular event reduction.

What role did progesterone play in ELITE?

Women with an intact uterus received vaginal micronized progesterone for 10 days each month. This was necessary for endometrial protection but introduces a potential confounding variable, since progesterone has independent vascular effects.

Who funded the ELITE trial?

Primary funding came from the National Institute on Aging (NIH). Supplemental support came from pharmaceutical sources. The principal investigator had a long research track record focused on estrogen and vascular biology.

Honest Criticisms and Limitations of the ELITE Trial

Q: Why did ELITE use carotid IMT instead of heart attack rates?

Measuring actual cardiovascular events would require tens of thousands of participants followed for 10+ years. CIMT allowed a smaller trial to test the timing hypothesis with a validated imaging biomarker. The tradeoff is that surrogate findings do not guarantee clinical event reduction.

Q: How does ELITE differ from the WHI?

WHI enrolled primarily older women (average age 63) and measured hard clinical outcomes over longer follow-up. ELITE enrolled both early and late postmenopausal women and used an imaging surrogate. The two trials address different questions at different timescales.

Q: Did the ELITE trial test transdermal estrogen?

No. ELITE tested only oral 17β-estradiol at 1 mg daily. Results cannot be directly extrapolated to transdermal patches, gels, or sprays, which have different pharmacokinetic and hepatic profiles.

Q: What was the dropout rate in ELITE?

Approximately 16% of participants discontinued the study drug before the end of the trial. The primary analysis used intention-to-treat methods, which included all randomized participants regardless of adherence.

Q: Did ELITE show any benefit in late-postmenopausal women?

No. Women who were 10 or more years past menopause showed no significant difference in CIMT progression between estradiol and placebo groups. The treatment-by-timing interaction was statistically significant.

Q: Has any trial replicated the ELITE finding?

No independent trial has replicated ELITE's primary result using the same design and population. KEEPS, which studied a similar age group, found no significant CIMT benefit with HRT, though it used different estrogen formulations and doses.

Q: Should doctors prescribe HRT for heart protection based on ELITE?

Current guidelines do not recommend initiating HRT solely for cardiovascular prevention. ELITE supports the safety of early HRT from a vascular standpoint but does not constitute proof of clinical cardiovascular event reduction.

Q: What role did progesterone play in ELITE?

Women with an intact uterus received vaginal micronized progesterone for 10 days each month. This was necessary for endometrial protection but introduces a potential confounding variable, since progesterone has independent vascular effects.

Q: Who funded the ELITE trial?

Primary funding came from the National Institute on Aging (NIH). Supplemental support came from pharmaceutical sources. The principal investigator had a long research track record focused on estrogen and vascular biology.

By HealthRX Medical Team

Published May 25, 2026Updated May 25, 2026Last reviewed May 25, 2026

At a glance

| Parameter | Detail | |-----------|--------| | N | 643 healthy postmenopausal women | | Intervention | Oral 17β-estradiol 1 mg/day (+ vaginal progesterone gel for those with a uterus) | | Comparator | Placebo | | Duration | Median 5 years | | Primary Endpoint | Rate of change in carotid artery intima-media thickness (CIMT) | | Key Result | CIMT progression significantly slower in early-postmenopausal group on estradiol vs placebo (p = 0.008); no benefit in late-postmenopausal group |

Why This Trial Matters Enough to Critique

The ELITE trial was designed to test the "timing hypothesis" of hormone therapy: that cardiovascular benefit depends on proximity to menopause onset. Its results were cited widely as confirmation that early HRT protects arteries while late HRT does not. The 2017 Hormone Therapy Position Statement from The North American Menopause Society explicitly referenced ELITE when supporting initiation of systemic estrogen in women under 60 or within 10 years of menopause.

Given the trial's influence on prescribing behavior and guideline language, its limitations deserve close inspection.

Enrollment and Selection Bias

Volunteer Effect

ELITE recruited exclusively from the greater Los Angeles area, primarily through advertisements. Women who respond to research ads tend to be healthier, more educated, and more health-aware than the general postmenopausal population. The published baseline characteristics confirm this: average BMI was approximately 27, smoking rates were low, and diabetes prevalence was minimal.

This creates a "healthy user" cohort that may respond to estrogen differently than typical clinical patients who carry more metabolic burden.

Ethnic Homogeneity

Approximately 60% of participants were non-Hispanic white. Hispanic women made up roughly 20%, with smaller representation from Black and Asian women. Cardiovascular disease risk profiles, estrogen metabolism, and arterial biology differ across ethnic groups. The trial was not powered to detect interaction effects by race, meaning the central finding may not generalize uniformly.

Exclusion of Higher-Risk Women

Women with clinical cardiovascular disease, uncontrolled hypertension, diabetes requiring medication, or BMI above 40 were excluded. These are precisely the patients for whom the timing hypothesis carries the most clinical stakes. ELITE tells us about estrogen's effect on healthy arteries, not diseased ones.

The Surrogate Endpoint Problem

CIMT Is Not a Hard Outcome

The primary endpoint was carotid intima-media thickness change measured by B-mode ultrasound. CIMT is a validated surrogate marker for atherosclerotic burden, but the relationship between CIMT changes and actual cardiovascular events (myocardial infarction, stroke, cardiovascular death) is not linear or guaranteed.

The METEOR trial of rosuvastatin demonstrated CIMT regression without proportional event reduction in low-risk patients. Similarly, several agents that improved CIMT in trials (niacin extended-release, for example) failed to reduce hard events in larger outcome trials.

ELITE was never designed or powered to detect differences in heart attacks or strokes. The secondary cardiac CT endpoint showed no significant difference in coronary artery calcium progression between groups, which adds ambiguity to the surrogate finding.

Measurement Variability

Ultrasound-based CIMT measurement is operator-dependent. ELITE used a centralized reading laboratory to minimize interobserver variability, which is good methodology. Still, the mean annual CIMT progression differences were small: approximately 0.0044 mm/year in the early group on placebo versus essentially no progression on estradiol. Differences of this magnitude approach the resolution limits of ultrasound technology and can be sensitive to measurement artifact.

Statistical Considerations

The Interaction Test

The primary publication reported a statistically significant interaction between time since menopause and treatment effect (p = 0.007 for the interaction term). This is the correct test for the timing hypothesis. Some critics, however, noted that the trial stratified by only two time categories (<6 years and ≥10 years post-menopause) rather than treating time as a continuous variable in the primary analysis.

A binary split at 6 years may oversimplify a biological gradient. Women at 5 years and 7 years post-menopause likely differ less from each other than the categorical analysis suggests.

Sample Size and Power

With 643 women split across four cells (early/late × estradiol/placebo), each cell contained roughly 160 participants. The trial was powered for the CIMT endpoint specifically and achieved its primary objective. But subgroup analyses within these cells (by age, BMI, baseline CIMT, adherence) are underpowered and should be interpreted cautiously.

Dropout and Adherence

Approximately 16% of participants discontinued study medication. The primary analysis used intention-to-treat methodology, which is conservative and appropriate. Per-protocol analyses were consistent with ITT findings, but selective dropout can still bias surrogate endpoint trials if healthier women preferentially remain on therapy.

Duration and Follow-Up Gaps

Five years of CIMT data provides a window into early atherosclerosis. It cannot address whether the early-group benefit persists after estrogen is stopped, accelerates, or translates into fewer clinical events over the next 10 to 20 years.

The Women's Health Initiative (WHI) follow-up data from the estrogen-alone arm suggested that the elevated risk signals observed during active treatment attenuated after stopping therapy. But WHI enrolled older women. Neither trial clarifies the long-term cardiovascular trajectory of starting HRT early and continuing it for decades, which is what many younger symptomatic women actually do.

Conflict-of-Interest Considerations

ELITE was funded by the National Institute on Aging with supplemental support from pharmaceutical sources. The principal investigator, Howard Hodis, has longstanding academic ties to estrogen research. While academic bias is not the same as commercial corruption, the research group had previously published the EPAT and WISH trials testing similar hypotheses with similar methods.

A research group deeply invested in the timing hypothesis may design, execute, and interpret data with subtle confirmation bias. This does not invalidate the findings, but it means independent replication carries extra weight. No fully independent trial has replicated ELITE's primary finding with the same design.

What Peer Commentary Raised

Following publication, several letters to the editor in the New England Journal of Medicine raised points including:

Clinical relevance: Slowing CIMT progression without demonstrated event reduction leaves prescribers in an uncertain position. Should a surrogate marker change prescribing behavior when WHI's hard-outcome data in older women showed harm?
Generalizability to transdermal routes: ELITE used oral estradiol exclusively. Transdermal estrogen avoids first-pass hepatic effects, has different impacts on clotting factors and inflammatory markers, and may have a distinct cardiovascular profile. ELITE's results cannot be assumed to apply to patches or gels.
Progesterone confounding: Women with an intact uterus received vaginal micronized progesterone for 10 days per cycle. Progesterone has independent vascular effects. The trial could not fully separate estrogen-only effects from combined regimen effects in this subgroup.
Absence of symptom data: ELITE did not systematically report hot flash reduction, sleep improvement, or quality of life. This limits the ability to weigh cardiovascular surrogate benefits against the symptomatic benefits that drive most HRT prescribing decisions.

What ELITE Cannot Tell Us

| Question | Why ELITE Cannot Answer It | |----------|---------------------------| | Does early HRT prevent heart attacks? | Surrogate endpoint only; no event power | | Is transdermal estrogen equally protective? | Only oral estradiol tested | | What happens after stopping HRT? | No post-treatment follow-up phase | | Do women with existing plaque benefit? | Excluded from enrollment | | Is the 6-year window a hard cutoff? | Binary stratification, not continuous modeling | | Does race modify the timing effect? | Underpowered for ethnic subgroups |

Placing ELITE in Context

The Kronos Early Estrogen Prevention Study (KEEPS) also studied early-postmenopausal women but found no significant CIMT difference between oral conjugated equine estrogens, transdermal estradiol, and placebo over 4 years. KEEPS used a different estrogen formulation at a lower dose, complicating direct comparison. The discordance between ELITE and KEEPS on the same surrogate endpoint in similar populations has not been fully reconciled.

The 2022 Menopause Society position statement continues to endorse HRT for symptomatic women in the early postmenopausal window, citing both ELITE and KEEPS alongside WHI subgroup analyses. Guidelines frame cardiovascular effects as "not harmful" in younger women rather than "proven beneficial," a nuance directly informed by ELITE's surrogate-only evidence base.

Bottom Line for Clinicians

ELITE is important mechanistic evidence supporting the timing hypothesis. It is not sufficient evidence to prescribe HRT primarily for cardiovascular protection. The trial enrolled healthy volunteers, measured a surrogate marker over a medium-term window, and cannot speak to hard outcomes, long-term trajectories, or populations with existing cardiovascular risk factors.

Clinicians should cite ELITE when reassuring early-postmenopausal patients that HRT is unlikely to accelerate atherosclerosis. They should not cite it as proof that estrogen prevents heart disease.

Frequently asked questions

›

References

Hodis HN, Mack WJ, Henderson VW, et al. Vascular Effects of Early versus Late Postmenopausal Treatment with Estradiol. N Engl J Med. 2016;374(13):1221-1231. PubMed
Harman SM, Black DM, Naftolin F, et al. Arterial imaging outcomes and cardiovascular risk factors in recently menopausal women: a randomized trial (KEEPS). Ann Intern Med. 2014;161(4):249-260. PubMed
The 2022 Hormone Therapy Position Statement of The North American Menopause Society. Menopause. 2022;29(7):767-794. PubMed
LaCroix AZ, Chlebowski RT, Manson JE, et al. Health outcomes after stopping conjugated equine estrogens among postmenopausal women with prior hysterectomy. JAMA. 2011;305(13):1305-1314. PubMed
Crouse JR 3rd, Raichlen JS, Riley WA, et al. Effect of rosuvastatin on progression of carotid intima-media thickness in low-risk individuals with subclinical atherosclerosis: the METEOR trial. JAMA. 2007;297(12):1344-1353. PubMed