Vaginal estrogen versus moisturizer in aromatase inhibitor users: randomized trial

For the millions of breast cancer survivors taking aromatase inhibitors, one of the most persistent and least discussed burdens of treatment has been the genitourinary syndrome of menopause—a cluster of symptoms including vaginal dryness, painful intercourse, and urinary discomfort that arises when estrogen is stripped from the body. Now, a randomized controlled trial known as VEMORA has delivered some of the first prospective randomized evidence that low-dose vaginal estrogen can safely and effectively relieve these symptoms in this vulnerable population, offering greater benefit than the non-hormonal moisturizers that have long been the conservative standard of care. The findings, published in Breast Cancer Research and Treatment, are poised to reshape a decades-long debate about whether topical estrogen has any place in the management of women with hormone receptor-positive breast cancer.

The clinical dilemma at the heart of the trial is rooted in the biology of modern breast cancer therapy. Approximately two-thirds of breast cancers are estrogen receptor-positive, meaning their growth is fueled by estrogen, and adjuvant endocrine therapy with aromatase inhibitors—anastrozole, letrozole, and exemestane—dramatically reduces recurrence risk by blocking the enzyme responsible for synthesizing estrogen from androgen precursors in peripheral tissues. In postmenopausal women, whose ovaries no longer produce meaningful quantities of estrogen, aromatase inhibitors can suppress circulating estradiol to nearly undetectable levels. That profound estrogen deprivation, however, comes at a physiological cost. The vaginal epithelium, which depends on estrogen to maintain thickness, elasticity, glycogen content, and an acidic protective pH, undergoes atrophy that produces dryness, irritation, and dyspareunia. Unlike hot flashes, which often improve over time, genitourinary symptoms tend to be progressive and unremitting, and they are among the leading reasons survivors abandon their lifesaving endocrine therapy prematurely.

Historically, clinicians have been caught between two imperfect options. Non-hormonal vaginal moisturizers, such as the polycarbophil-based product Replens, provide symptomatic hydration without any hormonal exposure and have been endorsed as the safest first-line choice. Vaginal estrogen, delivered as tablets, rings, or creams, is far more effective at reversing atrophy in the general postmenopausal population, but its use in breast cancer survivors has been controversial. Because systemic absorption cannot be entirely excluded, and because even small elevations in serum estradiol are theoretically undesirable in women whose tumors were estrogen-driven—and in whom aromatase inhibitors are specifically deployed to eliminate estrogen—the approach has been viewed with caution. A widely cited 2006 analysis argued that vaginal estradiol appeared contraindicated in women on aromatase inhibitors, and subsequent studies produced conflicting measurements of how much, if any, estradiol escapes into the circulation from low-dose vaginal preparations. Systematic reviews in recent years have consistently flagged the absence of randomized efficacy data as the critical gap.

VEMORA, conducted by investigators at Houston Methodist Hospital and Baylor College of Medicine and registered at ClinicalTrials.gov as NCT01984138, was designed to fill that gap. In this randomized, controlled, open-label trial, estrogen receptor-positive breast cancer survivors receiving adjuvant aromatase inhibitor therapy who suffered from symptomatic genitourinary syndrome of menopause were assigned to one of two treatments for 24 weeks. One group received a non-hormonal vaginal moisturizer, Replens, while the other received vaginal estrogen therapy in the form of either Vagifem vaginal tablets or the Estring vaginal ring, both of which release ultralow, continuous doses of estradiol locally. The primary endpoint was the change in vaginal dryness over the treatment period, with secondary endpoints encompassing dyspareunia, broader domains of sexual functioning, vaginal pH, and—critically for safety—serial serum estradiol measurements.

Twenty-three patients were enrolled, eleven randomized to vaginal estrogen and twelve to the non-hormonal arm. Small as the sample was, the signal was consistent and statistically meaningful. Vaginal dryness scores improved significantly more in the vaginal estrogen group than in the moisturizer group, with a p-value of 0.049. Dyspareunia, the pain associated with sexual activity that so often erodes intimacy and quality of life after breast cancer, likewise improved significantly more with estrogen, at p = 0.048. Perhaps the most mechanistically telling result involved vaginal pH, an objective biomarker of estrogenic activity in the genital tract. In healthy premenopausal women, estrogen-driven glycogen deposition in vaginal epithelial cells feeds lactobacilli, which produce lactic acid and maintain a pH of roughly 3.8 to 4.5. With estrogen loss, pH rises toward 6.0 or higher, the protective microbiome shifts, and symptoms follow. In VEMORA, vaginal pH decreased significantly with vaginal estrogen compared with non-hormonal therapy, at p = 0.0286, providing physiological confirmation that the locally administered estradiol was genuinely restoring the vaginal environment rather than merely lubricating the surface.

Not every dimension of sexual health moved in parallel. Domains including libido, the ability to achieve orgasm, overall sexual satisfaction, and the quality of the partner relationship did not differ significantly between the two groups. This pattern is not entirely surprising to researchers in sexual medicine, who note that desire and satisfaction are multifactorial constructs shaped by psychological wellbeing, body image after cancer treatment, partner dynamics, and fatigue, and may not track closely with tissue-level reversal of atrophy. Dryness and dyspareunia—the most directly mechanical and estrogen-dependent components of the genitourinary syndrome—are precisely where a local estrogen effect should be most visible, and that is exactly what the trial observed.

The safety data form the second half of the story and carry substantial implications. Serum estradiol levels remained low throughout the first 12 weeks of therapy in all participants, an encouraging finding given that the primary theoretical concern with vaginal estrogen in this population is systemic absorption potentially undermining aromatase inhibitor efficacy. Two participants, however, demonstrated estradiol elevations above a pre-specified threshold at the 24-week mark, reaching 21.8 pg/mL and 16 pg/mL respectively. Importantly, both of these individuals reported non-adherence to their aromatase inhibitor therapy in the period preceding the measurement. This detail is crucial to interpretation: without the aromatase inhibitor actively suppressing peripheral estrogen synthesis, endogenous estradiol production alone could account for the observed elevations, making it difficult or impossible to attribute those readings to the vaginal estrogen itself. The finding also serves as a reminder of the fragility of endocrine therapy adherence in real-world survivorship, where discontinuation rates approaching 30 to 50 percent over five years have been documented in large cohort studies, with consequences for mortality.

The measurement question adds another layer of technical nuance. Immunoassays for estradiol, which are widely used clinically, are known to suffer from cross-reactivity and limited sensitivity at the very low concentrations relevant to postmenopausal women on aromatase inhibitors, and comparisons between immunoassay and liquid chromatography-tandem mass spectrometry methods have revealed clinically meaningful discrepancies. Several of the estradiol-elevation reports that fueled anxiety about vaginal estrogen over the past two decades may reflect assay artifacts rather than true systemic exposure. The VEMORA investigators monitored estradiol serially and used a pre-defined threshold to flag concerning values, an approach that acknowledges both the assay limitations and the theoretical oncologic risk while allowing the question to be examined empirically rather than assumed.

What does VEMORA change in practice? The trial is, to date, one of the few—if not the first—randomized comparisons of vaginal estrogen against a non-hormonal moisturizer specifically in aromatase inhibitor users, and it provides prospective evidence that low-dose vaginal estrogen delivers symptomatic benefit with minimal detectable systemic estrogen exposure in most patients. Prior prospective work, including a phase II trial of an ultralow-dose 0.005% estriol vaginal gel and a 2025 prospective study of vaginal estrogen during aromatase inhibitor therapy published in the same journal, has moved in a similar direction, and a recent systematic review and meta-analysis of recurrence and mortality risks found no signal of harm. VEMORA’s randomized design strengthens this accumulating evidence base by directly comparing the two competing strategies head-to-head rather than benchmarking against historical controls.

The investigators and commentators alike are careful to note the limitations. With 23 participants, the study was powered to detect only large effects, the open-label design leaves room for expectancy effects in patient-reported outcomes, and 24 weeks of follow-up cannot address the long-term safety question—particularly recurrence risk over years—that ultimately matters most to oncologists and patients. The authors state explicitly that larger studies with longer follow-up are needed to define long-term safety and efficacy. Nonetheless, the trial shifts the framing of the conversation: rather than asking whether vaginal estrogen can ever be justified in these women, the question is becoming how to identify which patients, at what doses, with which monitoring strategies, can safely benefit.

For patients, the practical takeaway is nuanced but meaningful. The results suggest that vaginal estrogen should not be reflexively withheld from aromatase inhibitor users with bothersome genitourinary symptoms, and that the decision is best made through shared decision-making between survivor and oncology or gynecology team, weighing symptom severity against individual recurrence risk and preferences. For women who prefer to avoid any estrogen, effective non-hormonal options remain available, though VEMORA suggests they may be less effective for moderate to severe symptoms. The deeper significance may lie in what the trial represents for survivorship medicine more broadly: a recognition that the side effects of curative-intent therapy deserve rigorous, randomized study rather than conservative assumption, and that quality of life after breast cancer is not a luxury but an integral measure of treatment success.