Vaginal estrogen use in breast cancer survivors
Posted on July 21, 2025
Source:
American Journal of Obstetrics & Gynecology
Abstract
Objective
To assess the risk of breast cancer recurrence, breast cancer-specific mortality, and overall mortality for breast cancer survivors receiving vaginal estrogen therapy for genitourinary syndrome of menopause.
Data sources
From the inception of each database to April 6th, 2024, a systematic literature search was conducted in Google Scholar, PubMed, EMBASE, CINAHL, NCBI, and Science Direct. A secondary search was conducted on September 26th, 2024 utilizing Google Scholar, PubMed, EMBASE, CINAHL, and Science Direct.
Study eligibility criteria
We identified studies that reported on breast cancer recurrence defined per individual review criteria and considered both local and distant recurrence.
Study appraisal and synthesis methods
Three reviewers evaluated studies with eligibility criteria in mind. Breast cancer recurrence was the primary outcome. The secondary outcomes included: breast cancer mortality and overall mortality. Pooled unadjusted odds ratios with 95% confidence intervals were calculated using a random-effects model. We assessed the 95% prediction intervals to calculate the likely range within which we can expect to observe future individual values, based on a current model or dataset. We calculated the fragility index to evaluate the robustness of the pooled estimates.
Results
Of 5522 articles identified, 8 observational studies were included in this meta-analysis. The use of vaginal estrogen in patients with a history of breast cancer was not associated with an increased risk of breast cancer recurrence (6 articles, 24,060 patients, odds ratio, 0.48; 95% confidence interval, 0.23–0.98). There was no increase in the risk of breast cancer mortality (4 articles, 61,695 patients, odds ratio 0.60; 95% confidence interval 0.18–1.95). Lastly, there was no increase in overall mortality with use of vaginal estrogen in breast cancer survivors (5 articles 59,724, odds ratio 0.46; 95% confidence interval 0.42–0.49).
Conclusion
The use of vaginal estrogen in patients with a history of breast cancer does not appear to be associated with an increased risk of breast cancer recurrence, breast cancer-specific mortality, or overall mortality.
Introduction
Breast cancer is prevalent and treatment outcomes continue to improve.1 Accordingly, more than 4 million breast cancer survivors currently reside in the US.2 Many treatment methods for breast cancer can contribute to a patient developing genitourinary syndrome of menopause (GSM) which has become increasingly prevalent in breast cancer patients.1 GSM refers to atrophic genital and lower urinary tract changes which result from loss of estrogen. For instance, bilateral oophorectomy may be performed in premenopausal women with breast cancer. Chemotherapy can also be used to treat breast cancer in premenopausal women, which often induces menopause. Aromatase inhibitors, widely used to prevent recurrent breast cancer in menopausal women with hormone receptor positive tumors, exacerbate GSM.3 This condition often causes sexual dysfunction and impairs quality of life.3 Accordingly, obstetrician-gynecologists and others who provide care to adult patients routinely see patients with a personal history of breast cancer who are experiencing symptomatic GSM, often with accompanying sexual dysfunction.4
Why was this study conducted?
For breast cancer survivors, genitourinary syndrome of menopause (GSM) can greatly impair a woman’s quality of life. However, many clinicians are hesitant to prescribe vaginal estrogen to treat GSM in patients with a history of breast cancer due to fear of the patient developing recurrent disease.
Key findings
This systematic review and meta-analysis included a small number of observational studies assessing the association between vaginal estrogen use and breast cancer recurrence. The pooled data from these studies suggest that vaginal estrogen use is not significantly associated with an increased risk of breast cancer recurrence among survivors experiencing GSM.
What does this add to what is known?
This meta-analysis offers cautious reassurance regarding the safety of vaginal estrogen use in breast cancer survivors, based on the available observational data. While the findings may help providers feel more confident in considering off-label vaginal estrogen for this population, further research is needed to confirm these results. This could potentially improve the quality of life for breast cancer survivors, though clinical decisions should continue to be made on a case-by-case basis.
Vaginal estrogen is highly effective in treating symptomatic GSM.5 However, since the Food and Drug Administration (FDA) lists a personal history of breast cancer as a contraindication to use of all types of estrogens, including vaginal estrogen, safety concerns prevent many obstetrician-gynecologists and other clinicians from prescribing vaginal estrogen to breast cancer survivors.4 To assess the safety of vaginal estrogen in women with a history of breast cancer, we conducted a systematic review and meta-analysis of studies which have addressed the association between vaginal estrogen use and breast cancer recurrence, breast cancer-specific mortality and overall mortality.
Methods
Prior to conducting the literature searches, we registered the protocol with PROSPERO (ID: CRD42023479950) on September 12, 2023. This systematic review was conducted following the Cochrane Handbook for Systematic Reviews of Interventions and the 2020 Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, utilizing the Population, Intervention, Comparison, Outcomes and Study design (PICOS) framework to guide the search strategy.6,7
The population (P) for this study included women with GSM who have a history of breast cancer. The intervention (I) was the use of vaginal estrogen therapy, with comparators (C) such as placebo, no treatment, or alternative therapies (including nonhormonal treatments). The primary outcome (O) was the breast cancer recurrence rate, including local recurrence, contralateral breast cancer, or metastasis. The study designs (S) included observational studies, such as cohort and case-control studies, whether prospective or retrospective.
The search strategy was developed with the assistance of a medical librarian and ChatGPT (OpenAI, San Francisco, CA). The medical librarian facilitated access to databases such as Google Scholar, PubMed, ScienceDirect, CINAHL, and NCBI, while ChatGPT was used to craft the search terms based on the PICOS framework. Studies were not excluded based on language, publication year, or country of origin. The search keywords included terms such as “vaginal estrogen,” “urogenital atrophy,” “genitourinary syndrome,” and “breast neoplasm.” Given the specific focus on breast cancer recurrence, additional terms like “breast carcinoma” and “estrogen cream” were included, especially in relevant indexing fields.
Eligibility criteria, information sources, search strategy
We initially conducted a comprehensive literature search from inception to November 10, 2023, with the assistance of a medical librarian. The search was extended to April 6, 2024, to capture newly published studies. The databases searched included Google Scholar, PubMed, NCBI, ScienceDirect, EMBASE, and CINAHL. Three reviewers (M.B., A.M.K., J.M.) independently screened abstracts to exclude studies that clearly did not meet the inclusion criteria. For studies with potential relevance or those not definitively excluded in the initial screening, full-text articles were reviewed to determine final eligibility. Studies were included irrespective of language, country, or year of publication. A revised search, excluding 'outcomes' and focusing on relevant terms for vaginal estrogen treatment, was conducted on September 26th, 2024. This search did not yield any additional studies for inclusion. The full search syntaxes are available in the Appendix.
Inclusion criteria
Studies were eligible for inclusion if they examined the use of vaginal estrogen in any formulation among patients with a history of breast cancer, and if they compared outcomes to breast cancer patients who did not use vaginal estrogen. The studies had to report on at least one of the following outcomes: breast cancer recurrence, breast cancer-specific mortality, or overall mortality. Studies that assessed additional outcomes but still met these criteria were considered relevant for inclusion. We included observational studies, such as cohort and case-control studies (both prospective and retrospective).
Exclusion criteria
Studies were excluded if they focused on systemic hormone therapy with or without vaginal estrogen in breast cancer survivors. Additionally, studies that did not assess vaginal estrogen use in breast cancer survivors or did not report on the specific outcomes of interest were excluded from the study. After a full-text review, 13 articles were excluded. 6 studies were excluded based on the use of systemic hormone therapy with or without the use of vaginal estrogen.8–13 Four studies were excluded for no outcome of interest identified, primarily these studies evaluated estradiol levels and did not look at breast cancer recurrence, breast cancer mortality or overall mortality.14–17 Additionally, we excluded 3 studies due to repeating information.18–20
Study selection
Three reviewers (M.B., A.M.K., J.M.) independently screened abstracts to identify studies that potentially met the inclusion criteria. After the abstract screening, the full-text articles of selected studies were reviewed to confirm eligibility. Any discrepancies between the reviewers were resolved through discussion with the senior author (L.S.R.). In cases where additional data clarification was necessary, the team contacted 3 authors (McVicker, O'Meara, and Le Ray).21–23 McVicker and O'Meara were contacted to verify the data in a number of women instead of person-years. Le Ray was contacted for chart interpretation for final data analysis.
Data extraction
The primary outcome of interest in this study was the risk of breast cancer recurrence among breast cancer survivors in relation to the use or nonuse of vaginal estrogen. Secondary outcomes included breast cancer-specific mortality and overall mortality. Three reviewers (M.B., A.M.K., J.M.) independently extracted data, and any discrepancies were discussed and resolved with the senior author (L.S.R.). Microsoft Excel was used for data collection and sharing among the review team.
Assessment of quality of the included studies
To assess the quality of the included studies, 2 independent reviewers (M.B., J.M.) assigned Newcastle-Ottawa scores. Studies scoring between 7 and 9 were categorized as having a low risk of bias, those scoring 5 to 6 were considered to have a moderate risk of bias, and those scoring below 5 were considered to have a high risk of bias. Any discrepancies between the reviewers were resolved through discussion with the senior author (L.S.R.).
Data synthesis
For quantitative analyses, odds ratios (ORs) and 95% confidence intervals (CIs) were calculated for categorical outcomes using a random-effects model (DerSimonian and Laird).24
Statistical significance was set at a 2-sided P value of <.05. To evaluate the robustness of the summary effects, we calculated the fragility index (FI), which indicates the number of events that would need to change from a nonevent to an event (or vice versa) to shift the statistical significance of the results.25 A low FI indicates a more fragile study, while a high FI suggests greater robustness.25 For nonsignificant outcomes, the reverse fragility index (RFI) was calculated.
Heterogeneity among the studies was assessed using Tau-squared and Higgins’ I2 statistic, with substantial heterogeneity defined as an I2 value greater than 50%.26 Additional analyses were conducted to explore the sources of heterogeneity and to enhance the robustness of the findings. These included calculating 95% prediction intervals (PIs) to estimate the range within which the effect size of a future individual study is likely to fall. Sequential leave-one-out analyses were performed to evaluate the impact of each individual study on the overall summary estimates.
Since fewer than 10 studies assessed any given outcome, funnel plots were not constructed to evaluate publication bias or small-study effects. All meta-analyses were conducted using Stata version 17.0 (StataCorp LLC, College Station, TX).
Results
Study selection
The initial database searches yielded 5522 studies. Once we eliminated the duplicated studies or those with abstracts/titles not meeting inclusion criteria, 21 studies remained. After a full-text review, we excluded 13 additional studies. Finally, 8 studies met the inclusion criteria.21–23,27–31 Figure 1 illustrates the PRISMA flow diagram. Similar to the findings of a recent systematic review, we identified no randomized controlled trials that met the inclusion criteria.32