This cohort study compares the mortality risk associated with vaginal estrogen therapy vs no hormone replacement therapy in 2 cohorts of patients with breast cancer in Scotland and Wales.
Key Points
Question
Do females with breast cancer who use vaginal estrogen therapy, such as tablets or creams, have a higher risk of breast cancer–specific mortality?
Findings
In this cohort study of 49 237 females with breast cancer, there was no evidence of an increase in early breast cancer–specific mortality with use of vaginal estrogen therapy compared with no hormone replacement therapy use after breast cancer diagnosis.
Meaning
Findings of this study may provide some reassurance to clinicians and support the guidelines suggesting that vaginal estrogen therapy can be considered in patients with breast cancer and genitourinary symptoms if nonhormonal treatments were unsuccessful.
Abstract
Importance
Genitourinary syndrome of menopause can be treated with vaginal estrogen therapy. However, there are concerns about the safety of vaginal estrogen therapy in patients with breast cancer.
Objective
To determine whether the risk of breast cancer–specific mortality was higher in females with breast cancer who used vaginal estrogen therapy vs females with breast cancer who did not use hormone replacement therapy (HRT).
Design, Setting, and Participants
This cohort study analyzed 2 large cohorts, one each in Scotland and Wales, of females aged 40 to 79 years with newly diagnosed breast cancer. These population-based cohorts were identified from national cancer registry records from 2010 to 2017 in Scotland and from 2000 to 2016 in Wales and were followed up for breast cancer–specific mortality until 2020. Females were excluded if they had a previous cancer diagnosis (except nonmelanoma skin cancer). Data analysis was performed between August 2022 and August 2023.
Exposure
Use of vaginal estrogen therapy, including vaginal tablets and creams, was ascertained from pharmacy dispensing records of the Prescribing Information System for the Scotland cohort and from general practice prescription records for the Wales cohort.
Main Outcomes and Measures
The primary outcome was time to breast cancer–specific mortality, which was obtained from national mortality records. Time-dependent Cox proportional hazards regression models were used to calculate hazard ratios (HRs) and 95% CIs for breast cancer–specific mortality, comparing vaginal estrogen therapy users with HRT nonusers and adjusting for confounders, including cancer stage and grade.
Results
The 2 cohorts comprised 49 237 females with breast cancer (between 40 and 79 years of age) and 5795 breast cancer–specific deaths. Five percent of patients with breast cancer used vaginal estrogen therapy after breast cancer diagnosis. In vaginal estrogen therapy users compared with HRT nonusers, there was no evidence of a higher risk of breast cancer–specific mortality in the pooled fully adjusted model (HR, 0.77; 95% CI, 0.63-0.94).
Conclusions and Relevance
Results of this study showed no evidence of increased early breast cancer–specific mortality in patients who used vaginal estrogen therapy compared with patients who did not use HRT. This finding may provide some reassurance to prescribing clinicians and support the guidelines suggesting that vaginal estrogen therapy can be considered in patients with breast cancer and genitourinary symptoms.
Introduction
Many females with breast cancer experience symptoms of genitourinary syndrome of menopause,1 such as vaginal itchiness, burning, pain with sexual activity, and urinary incontinence. These symptoms may be precipitated by endocrine treatments and contribute to noncompliance with endocrine therapy.2 Vaginal estrogen therapy is an effective treatment for genitourinary syndrome of menopause.3 Trials have shown increased recurrence in patients with breast cancer who received systemic hormone replacement therapy (HRT).4 A recent trial observed a small increase in serum estradiol with use of a vaginal estradiol tablet (10 μg).5
There have been no large randomized clinical trials of vaginal estrogen therapy in patients with breast cancer that are powered to investigate recurrence or mortality,6 and observational studies have been limited by small sample size7,8 and unavailable confounders.9 A recent observational Danish study showed no increase in recurrence in patients with breast cancer who received vaginal estrogen therapy aside from a subgroup who received both vaginal estrogen therapy and aromatase inhibitors.10 Consequently, in this study, we investigated vaginal estrogen therapy and breast cancer–specific mortality in 2 large cohorts. We aimed to determine whether the risk of breast cancer–specific mortality was higher in females with breast cancer who used vaginal estrogen therapy vs females with breast cancer who did not use HRT.
Methods
We obtained data from the Prescribing Information System for the cohort in Scotland11 and from the SAIL Databank for the cohort in Wales.12 The SAIL Databank Information Governance Review Panel and the Privacy Advisory Committee of the National Health Service National Services Scotland approved this cohort study. Informed consent was not required as the data were deidentified prior to analysis. We followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.
Population-based cohorts of females aged 40 to 79 years with newly diagnosed breast cancer (International Statistical Classification of Diseases and Related Health Problems, Tenth Revision [ICD-10] code C50) were identified from cancer registries in Scotland from 2010 to 2017 and in Wales from 2000 to 2016. Patients who were previously diagnosed with other invasive cancers (except nonmelanoma skin cancer) were excluded.
Exposure, Outcome, and Covariates
Medication use was ascertained from general practitioner (GP) prescribing records for patients in Wales or from pharmacy dispensing records for patients in Scotland. Vaginal estrogen therapy (mainly estriol creams and estradiol tablets) and systemic HRT (including estrogen- or tibolone-containing products) were identified using the British National Formulary classification.13 The primary outcome of time to breast cancer–specific mortality was identified from national mortality records (an underlying cause of death for ICD-10 code C50) until June 2019 in Scotland and until June 2020 in Wales.
Cancer registry records provided stage, grade, treatment (radiotherapy, chemotherapy, or surgery), and, in Scotland, hormone receptor status. Tamoxifen, aromatase inhibitor, and other medication used were identified from prescribing or dispensing records. Charlson Comorbidity Index, anemia, and hysterectomy or oophorectomy were determined from GP diagnoses and hospital admissions in Wales and from hospital admissions alone in Scotland. Deprivation (defined as poverty level in an area) was based on the Index of Multiple Deprivation and the areas were categorized into fifths, with the highest fifth indicating the least deprived area. The GP records provided smoking status and body mass index data in Wales only.
Statistical Analysis
In the primary analysis (eFigure in Supplement 1), patients were followed up from 6 months after cancer diagnosis to breast cancer–specific mortality and censored, according to whichever occurred first, on death from other causes, end of mortality follow-up, or end of GP records in Wales and date of emigration in Scotland. The exposure was modeled as a single time-varying variable, with a lag of 6 months, into the following hierarchical categories: systemic HRT (with or without vaginal estrogen therapy), vaginal estrogen therapy alone, and no HRT. Analyses were conducted by number of prescriptions and separately for higher-dose vaginal estrogen therapy (consisting of 25-μg estradiol tablets).
Time-dependent Cox proportional hazards regression models were used to calculate hazard ratios (HRs) and 95% CIs by exposure, adjusting for age, year, deprivation level, cancer treatment (surgery, chemotherapy, and radiotherapy), tamoxifen or aromatase inhibitor use (as time-varying covariates with 6-month lags), Charlson Comorbidity Index (before diagnosis), anemia (before diagnosis), other medication use (including statins, aspirin, metformin, and oral contraceptives before diagnosis), hysterectomy or oophorectomy (anytime up to 6 months after diagnosis), and cancer stage and grade. Where missing, cancer stage and grade were imputed using multiple imputation with chained equations. Estimates were calculated within each cohort and pooled using random-effects meta-analysis models. The eMethods in Supplement 1 provide further details.
Two-sided P < .05 indicated statistical significance. Data analysis was performed between August 2022 and August 2023, using Stata, version 16/17 (StataCorp LLC).
Results
The 2 cohorts comprised 49 237 females with breast cancer (between 40 and 79 years of age) and 5795 breast cancer–specific deaths, with a median (IQR) duration of follow-up of 8 (5-12) years in the Wales cohort and 5 (3-7) years in the Scotland cohort. Overall, 5% of females (2551) used vaginal estrogen therapy after diagnosis and 1% (556) received systemic HRT.
Patient characteristics are shown in Table 1 and the eTable and eResults in Supplement 1. Table 2 shows there was no evidence of higher breast cancer–specific mortality in those who used vaginal estrogen therapy compared with those who used no HRT. A small decrease in mortality was found with vaginal estrogen therapy in the pooled fully adjusted model (HR, 0.77; 95% CI, 0.63-0.94). This estimate was similar in patients with 5 or more prescriptions (HR, 0.57; 95% CI, 0.34-0.96) and with higher-dose therapy (HR, 0.81; 95% CI, 0.55-1.21). Table 3 shows that the associations were similar in most sensitivity analyses. In particular, no increased risks were observed after restricting the model to females with estrogen receptor–positive breast cancer (HR, 0.88; 95% CI, 0.62-1.25) or females using aromatase inhibitors (HR, 0.72; 95% CI, 0.58-0.91). The eResults in Supplement 1 provide additional description of findings.
Table 1. Patient Characteristics by Hormone Replacement Therapy (HRT) Use After Diagnosis.
| Characteristic | Patients in Scotland, No. (%) | Patients in Wales, No. (%) | ||||
|---|---|---|---|---|---|---|
| No HRT | Systemic HRT | Only vaginal estrogen therapy | No HRT | Systemic HRT | Only vaginal estrogen therapy | |
| Age, y | ||||||
| 40-49 | 4207 (17) | 32 (15) | 184 (14) | 3491 (17) | 49 (14) | 184 (15) |
| 50-59 | 7444 (29) | 86 (39) | 455 (34) | 6143 (30) | 153 (45) | 411 (34) |
| 60-69 | 8231 (32) | 71 (33) | 436 (32) | 6685 (32) | 104 (31) | 394 (33) |
| 70-79 | 5506 (22) | 29 (13) | 281 (21) | 4423 (21) | 32 (9) | 206 (17) |
| Year of diagnosis | ||||||
| 2000-2004 | 0 | 0 | 0 | 4795 (23) | 139 (41) | 443 (37) |
| 2005-2009 | 0 | 0 | 0 | 6030 (29) | 94 (28) | 422 (35) |
| 2010-2014 | 15 674 (62) | 155 (71) | 1045 (77) | 6967 (34) | 86 (25) | 270 (23) |
| 2015-2017 | 9714 (38) | 63 (29) | 311 (23) | 2950 (14) | 19 (6) | 60 (5) |
| Deprivation level | ||||||
| First fifth: most deprived area | 5580 (22) | 44 (20) | 342 (25) | 3387 (16) | 70 (21) | 158 (13) |
| Fifth fifth: least deprived area | 4240 (17) | 38 (17) | 202 (15) | 4634 (22) | 73 (22) | 313 (26) |
| Hysterectomy or oophorectomya | ||||||
| Before or at cancer diagnosis | 1034 (4) | 23-28 (11)b | 50-55 (4)b | 1476 (7) | 41 (12) | 87 (7) |
| After cancer diagnosis | 740 (3) | <5b | 53 (4) | 1092 (5) | 33 (10) | 110 (9) |
| Select comorbidity, any time before diagnosis | ||||||
| COPD | 1413 (6) | 24 (11) | 90 (7) | 781 (4) | 23 (7) | 33 (3) |
| Diabetes | 1760 (7) | 12 (6) | 101 (7) | 1653 (8) | 21 (6) | 77 (6) |
| CKD | 250 (1) | <5b | 16 (1) | 1093 (5) | 8 (2) | 48 (4) |
| Anemia | 480 (2) | <5b | 33 (2) | 1135 (5) | 18 (5) | 55 (5) |
| Medication use, any time before diagnosis | ||||||
| Statin | 6254 (25) | 59 (27) | 361 (27) | 4920 (24) | 69 (20) | 263 (22) |
| Aspirin | 3742 (15) | 35 (16) | 213 (16) | 3360 (16) | 53 (16) | 174 (15) |
| Metformin | 1302 (5) | 8 (4) | 73 (5) | 1054 (5) | 18 (5) | 54 (5) |
| Oral contraceptive | 1666 (7) | 13 (6) | 83 (6) | 1841 (9) | 23 (7) | 90 (8) |
| Hormone receptor status | ||||||
| ER-positive | 21287 (84) | 171 (78) | 1136 (84) | NA | NA | NA |
| PR-positive | 14340 (57) | 136 (62) | 706 (52) | NA | NA | NA |
| ERBB2-positive | 3581 (14) | 25 (12) | 198 (15) | NA | NA | NA |
| Cancer stage | ||||||
| 1 | 11 150 (44) | 119 (55) | 710 (52) | 8475 (41) | 179 (53) | 554 (46) |
| 2 | 9513 (38) | 70 (32) | 490 (36) | 6812 (33) | 80 (24) | 331 (28) |
| 3 | 1903 (8) | 9 (4) | 65 (5) | 1698 (8) | 8-18b | 45-55b |
| 4 | 1183 (5) | 7 (3) | 21 (2) | 378 (2) | <10b | <10b |
| Missing data | 1639 (7) | 13 (6) | 70 (5) | 3379 (16) | 61 (18) | 255 (21) |
| Cancer grade | ||||||
| 1 | 3204 (13) | 39 (18) | 214 (16) | 3120 (15) | 66 (20) | 224 (19) |
| 2 | 11 899 (47) | 105 (48) | 680 (50) | 9390 (45) | 155 (46) | 535 (45) |
| 3 | 8827 (35) | 59 (27) | 406 (30) | 5205 (25) | 60 (18) | 266 (22) |
| Missing data | 1458 (6) | 15 (7) | 56 (4) | 3027 (15) | 57 (17) | 170 (14) |
| Cancer treatment | ||||||
| Surgery | 21 257 (84) | 196 (90) | 1234 (91) | 18 699 (90) | 304 (90) | 1110 (93) |
| Chemotherapy | 9393 (37) | 67 (31) | 465 (34) | 1500 (7) | 26 (8) | 85 (7) |
| Radiotherapy | 10 726 (42) | 95 (44) | 650 (48) | 6030 (29) | 63 (19) | 315 (26) |
| Hormonal treatment use, any time after diagnosis | ||||||
| Tamoxifen | 13 864 (55) | 109 (50) | 725 (54) | 12 721 (61) | 196 (58) | 690 (58) |
| Aromatase inhibitor | 12 191 (48) | 115 (53) | 769 (57) | 8722 (42) | 164 (49) | 648 (54) |
Abbreviations: CKD, chronic kidney disease; COPD, chronic obstructive pulmonary disease; ER, estrogen receptor; ERBB2, erb-b2 receptor tyrosine kinase 2 (formerly HER2); NA, not available; PR, progesterone receptor.
Hysterectomy or oophorectomy in the following periods: before cancer or at cancer diagnosis (anytime up to 6 months after cancer diagnosis) and after cancer diagnosis (>6 months after cancer diagnosis).
Range is shown to maintain statistical disclosure control.
Table 2. Vaginal Estrogen Therapy Use After Diagnosis and Breast Cancer–Specific Mortality .
| Analysis | No. of events | Person-years | Unadjusted HR (95% CI) | P value | Adjusted HR (95% CI)a | P value | Fully adjusted HR (95% CI)b | P value |
|---|---|---|---|---|---|---|---|---|
| Pooled | ||||||||
| No HRT | 5624 | 285 342 | 1 [Reference] | NA | 1 [Reference] | NA | 1 [Reference] | NA |
| Systemic HRT | 51 | 3894 | 0.75 (0.57-0.98) | .04 | 0.90 (0.63-1.28) | .56 | 0.98 (0.68-1.40) | .90 |
| Only vaginal estrogen therapy | 120 | 11437 | 0.66 (0.55-0.80) | <.001 | 0.72 (0.60-0.86) | <.001 | 0.77 (0.63-0.94) | .01 |
| 1-4 Vaginal estrogen therapy prescriptions | 105 | 9374 | 0.70 (0.58-0.85) | <.001 | 0.75 (0.62-0.92) | .005 | 0.81 (0.67-0.99) | .04 |
| ≥5 Vaginal estrogen therapy prescriptions | 15 | 2062 | 0.49 (0.30-0.82) | .007 | 0.55 (0.32-0.97) | .04 | 0.57 (0.34-0.96) | .03 |
| Lower-dose vaginal estrogen therapy | 92-97c | 9098 | 0.65 (0.53-0.80) | <.001 | 0.71 (0.55-0.93) | .01 | 0.77 (0.56-1.07) | .12 |
| Higher-dose vaginal estrogen therapyd | 23-28c | 2339 | 0.69 (0.39-1.21) | .20 | 0.78 (0.53-1.15) | .22 | 0.81 (0.55-1.21) | .31 |
| Scotland | ||||||||
| No HRT | 2293 | 115 520 | 1 [Reference] | NA | 1 [Reference] | NA | 1 [Reference] | NA |
| Systemic HRT | 15 | 859 | 0.91 (0.55-1.51) | .72 | 1.14 (0.69-1.90) | .61 | 1.26 (0.73-2.16) | .41 |
| Only vaginal estrogen therapy | 45 | 3979 | 0.65 (0.48-0.88) | <.001 | 0.78 (0.58-1.05) | .10 | 0.88 (0.65-1.19) | .40 |
| Wales | ||||||||
| No HRT | 3331 | 169 822 | 1 [Reference] | NA | 1 [Reference] | NA | 1 [Reference] | NA |
| Systemic HRT | 36 | 3035 | 0.69 (0.49-0.95) | .03 | 0.78 (0.56-1.09) | .15 | 0.86 (0.61-1.21) | .38 |
| Only vaginal estrogen therapy | 75 | 7458 | 0.67 (0.53-0.85) | .001 | 0.68 (0.54-0.86) | .001 | 0.71 (0.56-0.90) | .005 |
Abbreviations: HR, hazard ratio; HRT, hormone replacement therapy; NA, not applicable.
Model was adjusted for age, year, deprivation level, cancer treatment (surgery, radiotherapy, and chemotherapy), tamoxifen use (as time-varying covariate), aromatase inhibitor use (as time-varying covariate), Charlson Comorbidity Index (before diagnosis), anemia (before diagnosis), medication use (before diagnosis: statin, aspirin, metformin, and oral contraceptives), and hysterectomy or oophorectomy (before or at diagnosis).
Model included variables in adjusted model as well as imputed cancer stage and grade using multiple imputation.
Range is shown to maintain statistical disclosure control.
Higher-dose vaginal estrogen therapy consisted of 25-μg estradiol tablets, and lower dose consisted of all other vaginal estrogen therapy.
Table 3. Sensitivity Analyses for the Association Between Vaginal Estrogen Therapy Use and No Hormone Replacement Therapy (HRT) After Cancer Diagnosis.
| Analysis | No HRT events (person-years), No. | Vaginal estrogen therapy events (person-years), No. | HR (95% CI) | ||
|---|---|---|---|---|---|
| Unadjusted | Adjusteda | Fully adjustedb | |||
| Main analysis | 5624 (285 342) | 120 (11 437) | 0.66 (0.55-0.80) | 0.72 (0.60-0.86) | 0.77 (0.63-0.94) |
| With 1-y lag | 5132 (262 441) | 104 (10 202) | 0.67 (0.55-0.81) | 0.72 (0.59-0.87) | 0.77 (0.63-0.94) |
| With 2-y lag | 3932 (21 8204) | 76 (8046) | 0.63 (0.42-0.95) | 0.72 (0.57-0.90) | 0.75 (0.60-0.95) |
| Restricted to age 55-79 y at diagnosis | 3880 (187 722) | 86 (7745) | 0.67 (0.54-0.83) | 0.76 (0.61-0.95) | 0.82 (0.63-1.07) |
| Including age 18-79 y at diagnosis | 6062 (299 018) | 121 (11 725) | 0.64 (0.53-0.77) | 0.69 (0.57-0.82) | 0.74 (0.61-0.90) |
| Restricted to cancer stage I-III | 3551 (243 892) | 90 (9329) | 0.73 (0.59-0.90) | 0.75 (0.60-0.92) | 0.80 (0.65-0.99) |
| New HRTc | 5046 (233 546) | 68 (6572) | 0.66 (0.52-0.84) | 0.70 (0.55-0.90) | 0.76 (0.59-0.97) |
| Adjusting for prior HRT | 5624 (285 342) | 120 (11 437) | 0.66 (0.55-0.80) | 0.77 (0.64-0.92) | 0.81 (0.67-0.98) |
| ER-positive breast cancerd | 1516 (98 591) | 35 (33,66) | 0.69 (0.49-0.97) | 0.83 (0.59-1.16) | 0.88 (0.62-1.25) |
| ER-negative breast cancerd | 732 (15 438) | 10 (579) | 0.53 (0.28-0.98) | 0.55 (0.29-1.03) | 0.68 (0.36-1.28) |
| Stratifying entire cohorte | |||||
| No tamoxifen or aromatase inhibitor | 1752 (60 805) | 21 (2207) | 0.51 (0.33-0.78) | 0.56 (0.36-0.86) | 0.67 (0.43-1.04) |
| Tamoxifen only | 595 (88 062) | 14 (3433) | 0.86 (0.51-1.48) | 0.89 (0.52-1.53) | 1.01 (0.52-1.95) |
| Aromatase inhibitor, with or without tamoxifen | 3277 (13 6474) | 85 (5797) | 0.68 (0.54-0.84) | 0.70 (0.57-0.87) | 0.72 (0.58-0.91) |
| Stratifying only vaginal estrogen therapy usersf | |||||
| No tamoxifen or aromatase inhibitor | 5624 (285 342) | 21 (2207) | 0.61 (0.37-1.01) | 0.58 (0.36-0.94) | 0.68 (0.35-1.33) |
| Tamoxifen only | 5624 (285 342) | 14 (3433) | 0.26 (0.15-0.43) | 0.33 (0.20-0.56) | 0.41 (0.21-0.79) |
| Aromatase inhibitor, with or without tamoxifen | 5624 (285 342) | 85 (5797) | 0.94 (0.76-1.17) | 0.98 (0.79-1.22) | 0.99 (0.79-1.24) |
| Adjusting for cancer stage and grade, complete caseg | 3788 (231 575) | 88 (8886) | 0.71 (0.54-0.93) | 0.72 (0.54-0.94) | 0.82 (0.66-1.01) |
| Additionally adjusting for smoking status and BMI, multiple imputationg | 3331 (169 822) | 75 (7458) | 0.67 (0.53-0.85) | 0.68 (0.54-0.86) | 0.73 (0.57-0.92) |
| Breast cancer as any cause of death | 6489 (285 342) | 144 (11 437) | 0.68 (0.58-0.80) | 0.73 (0.62-0.86) | 0.77 (0.65-0.92) |
| Cardiovascular death | 919 (285 342) | 42-47 (11 437) | 0.80 (0.30-2.11) | 0.77 (0.28-2.15) | 0.78 (0.28-2.16) |
| All-cause mortality | 9612 (285 342) | 290 (11 437) | 0.73 (0.58-0.91) | 0.78 (0.69-0.88) | 0.80 (0.71-0.90) |
Abbreviation: BMI, body mass index (calculated as weight in kilograms divided by height in meters squared); ER, estrogen receptor; HR, hazard ratio.
Model was adjusted, except where otherwise stated, for age, year, deprivation level, cancer treatment (surgery, radiotherapy, and chemotherapy), tamoxifen use (as time-varying covariate), aromatase inhibitor use (as time-varying covariate), Charlson Comorbidity Index (before diagnosis), anemia (before diagnosis), medication use (before diagnosis: statin, aspirin, metformin, and oral contraceptives), and hysterectomy or oophorectomy (anytime before or up to 6 months after diagnosis).
Fully adjusted model consisted of, except where otherwise stated, variables in the adjusted model as well as imputed cancer stage and grade using multiple imputation.
Restricted to individuals who were not receiving HRT before breast cancer diagnosis.
Scotland only.
Stratifying entire cohort by endocrine therapy use (eg, vaginal estrogen therapy users who were not receiving tamoxifen or aromatase inhibitor were compared with HRT nonusers not receiving tamoxifen or aromatase inhibitors).
Stratifying only vaginal estrogen therapy users by endocrine therapy use, and hence the comparison group comprised all HRT nonusers in each analysis (eg, vaginal estrogen therapy users who were not receiving tamoxifen or aromatase inhibitor vs all HRT nonusers).
Wales only.
Discussion
In these large, contemporary population-based breast cancer cohorts, there was no evidence that vaginal estrogen therapy was associated with increased risk of early breast cancer–specific mortality. The null finding is similar to results of a Danish study involving 8461 patients with breast cancer that observed no association between vaginal estrogen therapy and cancer recurrence (adjusted HR, 1.08; 95% CI, 0.89-1.32).10 However, the Danish study observed a 39% increase in recurrence in users of both vaginal estrogen therapy and aromatase inhibitors.10 We did not study recurrence, but we observed no evidence of an increase in breast cancer–specific mortality in this subgroup. Additionally, a case-control study showed no association between vaginal estrogen therapy and breast cancer recurrence (identified from GP records) among tamoxifen users but did not adjust for cancer stage.9 Two small cohort studies found no increase in cancer recurrence in patients with breast cancer who received vaginal estrogen therapy,7,8 but both studies included fewer than 10 recurrences in the exposed group. A recent Swedish case-control study showed no increase in breast cancer–specific mortality in patients with breast cancer who used estrogen but did not distinguish between vaginal or systemic estrogen.14
In the absence of trials of vaginal estrogen therapy in breast cancer, the findings of this study provide some reassurance that patients with breast cancer who received vaginal estrogen therapy were not at a markedly higher risk of breast cancer–specific mortality and appear to support national guidelines suggesting that vaginal estrogen therapy can be considered for genitourinary symptoms if nonhormonal treatments are unsuccessful.3,15 The systemic HRT associations were examined for completeness but should not be a factor in clinical decisions given the wide CIs and previous trial observations of increased risks of recurrence with systemic HRT.4
Strengths and Limitations
Strengths of this study were the large population-based cohorts with up to 20 years of follow-up with linked prescribing or dispensing records, which eliminated recall bias and captured all HRT prescriptions. However, the study had limitations, including our inability to confirm medication adherence. The duration of follow-up did not allow the investigation of later breast cancer–specific mortality, and thus further research with extended follow-up is recommended. We adjusted for many important confounders, including cancer stage and grade and, in a sensitivity analysis, body mass index and smoking status, but we could not rule out residual confounding from poorly recorded or unavailable variables, such as physical activity and menopausal status.3 Estrogen receptor status of the tumor was not complete, but results were similar in endocrine therapy users, who have estrogen receptor–positive disease. Moreover, patients who received treatment for genitourinary syndrome of menopause may have lower estradiol levels and/or better compliance to endocrine therapies and thus have lower breast cancer–specific mortality.
Conclusions
In this large cohort study, there was no evidence of increased early breast cancer–specific mortality in females with breast cancer who received vaginal estrogen therapy compared with females with breast cancer who had no HRT. This finding may provide some reassurance to prescribing clinicians and support the guidelines suggesting that vaginal estrogen therapy can be considered in patients with breast cancer and genitourinary symptoms if nonhormonal treatments are unsuccessful.
eMethods.
eResults.
eTable. Additional Patient Characteristics by HRT Use After Diagnosis in Patients
eFigure. Figure Illustrating the Study Design for the Main Analysis of Vaginal Estrogen Therapy
eReferences
Data Sharing Statement
References
- 1.Baber RJ, Panay N, Fenton A; IMS Writing Group . 2016 IMS recommendations on women’s midlife health and menopause hormone therapy. Climacteric. 2016;19(2):109-150. doi: 10.3109/13697137.2015.1129166 [DOI] [PubMed] [Google Scholar]
- 2.Sousa MS, Peate M, Jarvis S, Hickey M, Friedlander M. A clinical guide to the management of genitourinary symptoms in breast cancer survivors on endocrine therapy. Ther Adv Med Oncol. 2017;9(4):269-285. doi: 10.1177/1758834016687260 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.“The 2022 Hormone Therapy Position Statement of The North American Menopause Society” Advisory Panel . The 2022 hormone therapy position statement of The North American Menopause Society. Menopause. 2022;29(7):767-794. doi: 10.1097/GME.0000000000002028 [DOI] [PubMed] [Google Scholar]
- 4.Poggio F, Del Mastro L, Bruzzone M, et al. Safety of systemic hormone replacement therapy in breast cancer survivors: a systematic review and meta-analysis. Breast Cancer Res Treat. 2022;191(2):269-275. doi: 10.1007/s10549-021-06436-9 [DOI] [PubMed] [Google Scholar]
- 5.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. doi: 10.1001/jamanetworkopen.2022.41743 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Biglia N, Bounous VE, D’Alonzo M, et al. Vaginal atrophy in breast cancer survivors: attitude and approaches among oncologists. Clin Breast Cancer. 2017;17(8):611-617. doi: 10.1016/j.clbc.2017.05.008 [DOI] [PubMed] [Google Scholar]
- 7.Dew JE, Wren BG, Eden JA. A cohort study of topical vaginal estrogen therapy in women previously treated for breast cancer. Climacteric. 2003;6(1):45-52. doi: 10.1080/cmt.6.1.45.52 [DOI] [PubMed] [Google Scholar]
- 8.O’Meara ES, Rossing MA, Daling JR, Elmore JG, Barlow WE, Weiss NS. Hormone replacement therapy after a diagnosis of breast cancer in relation to recurrence and mortality. J Natl Cancer Inst. 2001;93(10):754-762. doi: 10.1093/jnci/93.10.754 [DOI] [PubMed] [Google Scholar]
- 9.Le Ray I, Dell’Aniello S, Bonnetain F, Azoulay L, Suissa S. Local estrogen therapy and risk of breast cancer recurrence among hormone-treated patients: a nested case-control study. Breast Cancer Res Treat. 2012;135(2):603-609. doi: 10.1007/s10549-012-2198-y [DOI] [PubMed] [Google Scholar]
- 10.Cold S, Cold F, Jensen MB, Cronin-Fenton D, Christiansen P, Ejlertsen B. Systemic or vaginal hormone therapy after early breast cancer: a Danish observational cohort study. J Natl Cancer Inst. 2022;114(10):1347-1354. doi: 10.1093/jnci/djac112 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Alvarez-Madrazo S, McTaggart S, Nangle C, Nicholson E, Bennie M. Data resource profile: the Scottish National Prescribing Information System (PIS). Int J Epidemiol. 2016;45(3):714-715f. doi: 10.1093/ije/dyw060 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Ford DV, Jones KH, Verplancke JP, et al. The SAIL Databank: building a national architecture for e-health research and evaluation. BMC Health Serv Res. 2009;9:157. doi: 10.1186/1472-6963-9-157 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Joint Formulary Committee . British National Formulary 80: September 2020-March 2021. BMJ Group and Pharmaceutical Press; 2020. [Google Scholar]
- 14.Sund M, Garmo H, Andersson A, Margolin S, Ahlgren J, Valachis A. Estrogen therapy after breast cancer diagnosis and breast cancer mortality risk. Breast Cancer Res Treat. 2023;198(2):361-368. doi: 10.1007/s10549-023-06871-w [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.American College of Obstetricians and Gynecologists’ Committee on Clinical Consensus-Gynecology . Treatment of urogenital symptoms in individuals with a history of estrogen-dependent breast cancer: clinical consensus. Obstet Gynecol. 2021;138(6):950-960. doi: 10.1097/AOG.0000000000004601 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
eMethods.
eResults.
eTable. Additional Patient Characteristics by HRT Use After Diagnosis in Patients
eFigure. Figure Illustrating the Study Design for the Main Analysis of Vaginal Estrogen Therapy
eReferences
Data Sharing Statement