The Risk of Cardiovascular Disease in Postmenopausal Oestrogen Receptor-Positive Breast Cancer Patients Treated with Aromatase Inhibitors: A UK Population-based Study

Background: Over 80% of breast cancers are oestrogen receptor (ER) positive and treated with endocrine therapies including tamoxifen and aromatase inhibitors (AIs) to reduce the risk of recurrence. However, there are concerns about the cardiovascular safety of AIs.

Objectives: To determine whether new users of AIs are at an increased risk of cardiovascular disease (CVD) compared to new-users of tamoxifen.

Methods: In a population-based cohort study, we identified postmenopausal women newly diagnosed with non-metastatic ER-positive breast cancer between 2010 and 2017 from the Scottish Cancer Registry. Those with first-time dispensed prescriptions for tamoxifen or an AI within 6 months (identified from the Prescribing Information System) were included. Any CVD event, consisting of a hospitalisation or death from CVD (identified from hospital and death records respectively) was investigated. Follow-up began at prescription date and ended at the earliest date of a CVD event, death, leaving Scotland or the end of the study period (31st May 2019). Cox regression was used to calculate hazard ratios (HRs) and 95% confidence intervals (CIs) for CVD risk among AI users compared to tamoxifen users adjusting for confounding using inverse probability of treatment weighting. Secondary outcomes included specific types of CVD including myocardial infarction, stroke, heart failure, venous thromboembolism, angina and arrhythmias.

Results: There were 10,376 breast cancer patients included in the analysis, 6,166 of whom initiated therapy with an AI and 4,210 of whom initiated with tamoxifen. Among AI and tamoxifen users there were 929 and 459 CVD events, respectively. In the main analysis AI use was not associated with an increased risk of CVD (HR 1.10, 95% CI 0.92-1.32). However, among those who used AIs for at least 1 year, there was a 23% increased risk of CVD compared to tamoxifen (HR 1.23, 95% CI 1.01-1.50). In secondary analyses the risk of heart failure was elevated for AI users (HR 1.79, 95% CI 1.04-3.08) compared to tamoxifen users, though the number of events were small.

Conclusions: Our preliminary results suggest that longer-term use of AIs may be associated with an increased risk of CVD. Additional analyses will be conducted to quantify the absolute differences in CVD risk between AI and tamoxifen users.