Excerpts from the World Medical Literature: Gynaecology

Agrawal P, Singh SM, Able C, et al. Safety of vaginal estrogen therapy for genitourinary syndrome of menopause in women with a history of breast cancer. Obstet Gynecol 2023;142:660–8. Summary: This was an analysis of a US health research network from January 2009 to June 2022, involving patients diagnosed with breast cancer 3 months to 5 years before an initial diagnosis of genitourinary syndrome of menopause (GSM). Breast cancer and GSM diagnoses were from International Classification of Diseases (ICD) codes. Two groups were compared: a study cohort that had at least 3 prescriptions for vaginal estrogen (either estradiol or conjugated estrogens) with at least one prescription <1 year of the GSM diagnosis; and a control cohort that had no prescriptions for vaginal estrogen <1 year of the GSM diagnosis. Matching was done 1:1 via propensity scores. Importantly, the authors excluded those with active breast cancer, which they defined as the following events occurring <3 months from the diagnosis of GSM: mastectomy, radiation, and chemotherapy. Another exclusion was the development of another cancer 5 years before or after the diagnosis of GSM. For the outcome, the authors defined recurrence as follows: mastectomy, radiation, chemotherapy, or a secondary malignancy (metastatic disease) from 3 months to 5 years after vaginal estrogen was started. There was a total of 42 113 subjects in the study with GSM diagnosis after breast cancer; a minority of which had ductal carcinoma in situ (1003). 5% (2011) of the total subjects met the criteria for the study cohort; the control cohort consisted of a similarly sized sample that was propensity score matched. For all breast cancers, the risk of recurrence was not different between the study (vaginal estrogen) cohort (17.6%) and control cohort (17.1%) (risk ratio (RR) = 1.03, 95% CI = 0.91–1.18). There were, further, no differences in specific definitions of recurrence (mastectomy, chemotherapy, radiation). Interestingly, the study group had a lesser risk of secondary malignancy (metastatic disease) (RR = 0.52, 95% CI = 0.37–0.72). There were also no differences in recurrence-free survival or all-cause mortality at 5 years. The study group had a lower all-cause mortality at 10 years (RR = 0.70, 95% CI = 0.51–0.96). Three in 10 (30%) of the total cohort (12 620) had estrogen receptor status data: 10 584 were estrogen receptor-positive and 2036 were estrogen receptor-negative. Those with estrogen receptor-positive breast cancers were less likely to receive vaginal estrogen. Amongst subjects with estrogen receptor-positive breast cancer, there were no differences in recurrence between the vaginal estrogen study group and the control group (RR = 0.94, 95% CI = 0.77–1.15). Similarly, there were no differences in recurrence-free survival. Another sub-analysis was done comparing individuals with vaginal estrogen and aromatase inhibitor usage, contrasted to only vaginal estrogen. The group with both vaginal estrogen and aromatase inhibitor had a higher risk of recurrence and recurrence-free survival, both for all cancers and for estrogen receptor-positive cancers specifically. Comment: The authors concluded that in a claims-based study, they did not find an elevated risk of recurrence of breast cancer in those who were taking vaginal estrogen for GSM. They summarize the scope of the clinical problem, by noting the majority of breast cancer survivors post-menopause who undergo systemic treatment develop GSM, and the concern about the degree of vaginal estrogen systemic absorption in those with a history of breast cancer particularly if the cancer was estrogen receptor-positive. All in all, the findings of this study are encouraging about the option of vaginal estrogen to treat GSM in patients with a history of breast cancer, regardless of estrogen receptor status. There are a few caveats. First, it is critical to emphasize that the study excluded those with active disease—defined as <3 months from the diagnosis of GSM, the occurrence of mastectomy, radiation or chemotherapy. The results of this study do not apply to this special population. Second, they also noted their findings of increased risk of recurrence amongst those taking both vaginal estrogen and aromatase inhibitors are consistent with the literature, and that the explanation was unclear and may be multifactorial. Given this, it may be advisable to avoid vaginal estrogen in those taking aromatase inhibitors for history of breast cancer, and there remains a need for treatment options for these patients. McVicker L, M Labeit A, Coupland CAC, et al. Vaginal estrogen therapy use and survival in females with breast cancer. JAMA Oncol 2024;10:103–8. Summary: This was another observational analysis of 2 large cohorts from Scotland and Wales. Inclusion criteria were 40–79 years old with a new diagnosis of breast cancer from 2010–2017 in Scotland and 2000–2016 in Wales, using ICD codes. Vaginal estrogen use was defined as tablets or creams and were primarily estriol creams and estradiol tablets. Follow-up was done until 2019–2020 for the primary outcome of breast cancer-specific mortality, with a Cox regression analysis for time to mortality. Adjustments were made for potential confounders such as stage and grade. The total sample size was 49 237, amongst which there were 5795 breast cancer-specific deaths. Comparing those who used vaginal estrogen to those that did not, and after adjustments, the hazard ratio was 0.77 (95% CI = 0.63–0.94) which represented a small decrease in risk for those who used vaginal estrogen. Similarly, there was no increased risk with ≥5 prescriptions or higher estrogen doses (e.g., 25 mcg tablets). No major differences were seen in sensitivity analyses, including no increase in risk (hazard ratio = 0.88, 95% CI = 0.62–1.25) in cases of estrogen receptor-positive breast cancer. Comment: The authors concluded that vaginal estrogen did not correlate to higher early breast cancer-specific mortality. In their introduction, they say that systemic hormone therapy has been associated with increased breast cancer recurrence and that a recent study (see third paper discussed below) found a slight increase in serum estradiol with vaginal estradiol 10 mcg tablets. Given their data did not show a higher risk of mortality after vaginal estrogen, the authors point out that their findings support guidelines for vaginal estrogen for GSM after non-hormonal therapies have been trialled and ineffective. However, limitations with these types of administrative data cohorts include the issues of residual confounding and medication adherence (vs. only tracking prescriptions). Ultimately, the issue with both these studies is that they are observational. What is needed is a randomized controlled trial (RCT), particularly one with a large enough sample size to examine recurrence (or mortality) as an outcome. In the absence of such an RCT of vaginal estrogen to treat GSM in patients with prior breast cancer, these observational data do provide some reassurance to patients and their care providers. Mitchell CM, Larson JC, Crandall CJ, et al. Association of vaginal estradiol tablet with serum estrogen levels in women who are postmenopausal: secondary analysis of a randomized clinical trial. JAMA Netw Open 2022;5(11):e2241743. Summary: This was a post-hoc study of an RCT, to examine serum estrogen level with vaginal estrogen 10 mcg tablet daily for 2 weeks and then twice a week (plus a placebo gel), compared to a dual placebo, in postmenopausal patients after 12 weeks. There were 174 patients in the study: 88 with estrogen and 86 in the placebo group. The estrogen group had higher serum estradiol (geometric means 4.3 pg/mL vs. 3.5 pg/mL, P = 0.01), including after adjustment for factors such as pre-treatment level, age, and BMI (the latter being associated with higher serum estrogen). In a sub-analysis of those with estradiol <2.7 pg/mL at baseline, the same proportions (38.1% vs. 34.4%) ended up with serum estradiol >2.7 pg/mL for the estrogen and placebo groups (2.7 pg/mL being a potential cut-off denoting increased risk of breast cancer). No differences in estrone and sex hormone-binding globulin (SHBG) were observed. Comment: The authors concluded, based on this study and previous research, that any changes in serum levels after vaginal estrogen "likely have limited clinical significance". As estradiol is associated with alterations in estrone and SHBG, but these factors were not increased in the study, the authors stated that the increase in serum estradiol in this study was unlikely to have significant physiologic impacts. Together with the observational data above, this study provides additional reassurance about the use of vaginal estrogen to treat GSM in the context of prior breast cancer. That being said, this would still be a decision between each patient and their clinician. For some, the finding of some elevation in serum estrogen (even if not considered clinically significant) may not be acceptable. For others, including perhaps those with severe GSM signs/symptoms, these serum estrogen data may make them more comfortable proceeding with vaginal estrogen. Li FG, Fuchs T, Deans R, et al. Vaginal epithelial histology before and after fractional CO2 laser in postmenopausal women: a double-blind, sham-controlled randomized trial. Am J Obstet Gynecol 2023;229:278.e1–9. Summary: This paper was a histological study within a single-center, double-blind, sham-controlled RCT. The cohort consisted of 49 subjects (46 analyzed) who were postmenopausal and had at least one symptom of GSM for which they had sought medical help and not previously had an energy-based treatment, and where medical therapies had been previously trialled, declined, or otherwise contraindicated. Any previous vaginal estrogen use had to stop 6 months before the start of the trial. Exclusion criteria were active genital infection. Subjects were randomized to laser treatment or sham. For the CO2 laser, 3 vaginal fractional micro-ablative treatments were done 4–8 weeks apart (MonaLisa Touch), at what was described as standard settings. For the sham, minimal energy settings were used with no tissue effect. Primary outcome was vaginal wall biopsies that were taken before and after treatment, and that were pathologically classified into well-estrogenized, poorly estrogenized, or combination. These biopsies were 3 cm above the hymenal ring at the right lateral vaginal wall at baseline and left lateral vaginal wall after 6 months of the laser or sham treatment. Pathologists were specialist gynaecologic pathologists who were blinded. The classification was based on glycogen-rich squamous epithelium, vascularization of the sub-epithelial stroma, and thickness of the basal epithelium, and was found to be reproducible. Subjects were classified according to whether the classification had progressed (worsened), unchanged, or regressed (improved). Other outcomes were subjective symptoms on a visual analogue scale, the Vulvovaginal Symptom Questionnaire, and Vaginal Health Index. The authors found no differences in the pathologic assessment of the vaginal mucosa between laser treatment and the sham. Subgroup analyses showed the same finding. Interestingly, the pathologic classification did not show a strong association with patient symptoms of GSM. Comment: The authors concluded that "fractional CO2 laser is not significantly different from sham treatment and should not be recommended for clinical use for postmenopausal vaginal symptoms". They summarized the literature and remarked that it was contradictory, with some studies showing similar findings to this trial, and others seemingly showing benefits to energy-based therapies. They commented that the latter studies tended to be of small sample size and with non-blinding of pathologists. This histological study is in line with the lack of differences in patient or clinician-reported outcomes after 12 months that were previously reported by the authors for the laser versus sham.1Li F.G. Maheux-Lacroix S. Deans R. et al.Effect of fractional carbon dioxide laser vs sham treatment on symptom severity in women with postmenopausal vaginal symptoms: a randomized clinical trial.JAMA. 2021; 326: 1381-1389Google Scholar Together, these results do not provide evidence for the use of vaginal fractional CO2 laser as a non-hormonal treatment for GSM, which would be of particular interest to patients with a history of breast cancer. It should be noted that there are variations in energy-based treatments, and the results of this trial should not necessarily be extrapolated to these other options. However, the randomized trial evidence from this group would seemingly reduce enthusiasm for this overall treatment approach to GSM. Fortunately, the vaginal estrogen data summarized above would suggest that it remains a feasible treatment for the majority of patients with GSM, including some who have a history of breast cancer. Extraits de la littérature médicale mondiale : gynécologieJournal of Obstetrics and Gynaecology Canada Vol. 46Issue 4PreviewAgrawal P, Singh SM, Able C, et al. Safety of vaginal estrogen therapy for genitourinary syndrome of menopause in women with a history of breast cancer. Obstet Gynecol 2023;142:660–8.