To Örebro University

IMPORTANCE: Endocrine treatments, such as Tamoxifen (TAM) and/or Aromatase inhibitors (AI), are the adjuvant therapy of choice for hormone-receptor positive breast cancer. These agents are associated with menopausal symptoms, adversely affecting drug compliance. Topical estrogen (TE) has been proposed for symptom management, given its' local application and presumed reduced bioavailability, however its oncological safety remains uncertain.

OBJECTIVE: The present systematic review, meta-analysis and expert panel review aimed to evaluate the strength of the available evidence on the risk of recurrence and mortality when TE is utilised in congruence with TAM or AI treatment, among BC survivors.

DATA SOURCES: Six databases and two prospective registers, were interrogated from inception to January 3rd, 2024. Search terms were Breast cancer AND Hormone replacement therapy AND topical/vaginal oestrogen AND recurrence/mortality.

STUDY SELECTION: All study designs reporting the use vs. non-use of TE in breast cancer survivors receiving adjuvant endocrine treatment were included. Six observational studies were deemed eligible for inclusion. DATA EXTRACTION AND SYNTHESIS: Sources of heterogeneity were explored using subgroup analysis by risk of bias, median follow-up period, node positivity and menopausal status. Trial sequential analysis was performed to quantify outcome reliability. A global expert panel was called to deliberate on the data, pinpoint areas of limited understanding, and determine the most important areas for future research.

MAIN OUTCOMES AND MEASURES: Risk ratio effect sizes (RR) and corresponding 95 % Confidence Intervals (CI) of breast cancer recurrence and mortality in survivors on endocrine treatment (TAM and/or AI) exposed to TE were reported. Expert panel appraisal of meta-analysis evidence with definition of current knowledge gaps and future research aims.

RESULTS: In 38 050 female patients receiving adjuvant endocrine treatment, of whom 1805 had been exposed to TE, TE exposure of those on AI, did not increase all-cause mortality (RR 0.99 [95 %CI 0.58, 1.69], I2 = 81 %, P = 0.96; moderate GRADE certainty). However, such exposure may convey an increased risk of recurrence (RR 2.51 [95 % CI 1.10, 5.72], I2 = 9 %, P = 0.03; low-GRADE certainty). Exposure to TE during TAM did not increase either recurrence risk or all-cause mortality. Clinical factors such as lymph node positivity at the time of diagnosis and menopausal status and follow-up time appeared to be significant confounders.

CONCLUSIONS AND RELEVANCE: The use of TE does not appear to increase either recurrence or mortality risk among BC survivors treated with TAM. An increased recurrence risk, without an increase in mortality, cannot be ruled out when TE is used during AI.