Ischaemic cardiotoxicity of aromatase inhibitors in postmenopausal patients with early breast cancer in Denmark: a cohort study of real-world data

Abstract

Background

For aromatase inhibitor treatment (AIT) in breast cancer, there is an unresolved concern about ischaemic cardiotoxicity. We investigated the association between AIT and ischaemic cardiotoxicity in a prospective cohort of female patients with early breast cancer who received contemporary treatment in Denmark.

Methods

In this prospective cohort study in Denmark, we identified postmenopausal patients of any age diagnosed with breast cancer as recorded in the nationwide Danish Breast Cancer Cooperative Group (DBCG) clinical database between Jan 1, 2009, and Dec 31, 2020, and linked them to other nationwide registries. Exclusion criteria included having a history of other primary cancer, less than 2 years of residency in Denmark, and no inclusion in a treatment protocol according to the DBCG database, including for metastatic or locally advanced breast cancer. Information on demography, hospital diagnoses, filled prescriptions, laboratory testing, and socioeconomic status were recorded. We stratified the patient cohort according to history (yes vs no) of selected cardiovascular disease defined as ischaemic heart disease, ischaemic stroke, and heart failure, and defined the primary outcome as two-point major adverse cardiovascular events (MACE; acute myocardial infarction or ischaemic stroke). We estimated cause-specific hazard ratios (HRs) according to allocation to AIT versus not in an intention-to-treat analysis using a Cox proportional hazards regression model with age as the underlying time scale, adjusting for demographic characteristics, tumour characteristics, and other anti-cancer treatments.

Findings

43 440 postmenopausal patients diagnosed with breast cancer were identified, of whom 32 635 were followed up and included in analyses. Of 29 118 postmenopausal patients with no history of selected cardiovascular disease, we observed 510 two-point MACEs among 22 135 patients allocated to AIT (incidence rate 4·3/1000 person-years of follow-up) and 170 two-point MACEs among 6983 patients not allocated to AIT (4·1/1000 person-years). The adjusted HR was 0·91 (95% CI 0·73–1·14) for patients allocated to AIT versus patients not allocated to AIT. Among 3517 patients with a history of selected cardiovascular disease, we observed 158 two-point MACEs among 2661 patients allocated to AIT (incidence rate 12·4/1000 person-years) and 50 two-point MACEs (12·1/1000 person-years) among 856 patients not allocated to AIT (adjusted HR 0·81 [95% CI 0·58–1·15]).

Interpretation

Our findings do not support a clinically relevant ischaemic cardiotoxic potential of AIT in patients with early breast cancer and do not support avoiding AIT prescription in patients with early breast cancer.

Funding

Bispebjerg and Frederiksberg Hospital, Kræftens Bekæmpelse, Fonden til Lægevidenskabens Fremme, Aase og Ejnar Danielsens Fond, Helsefonden, and Læge Sofus Carl Emil Friis og Hustru Olga Doris Friis’ Legat.