Endocrine therapy use and the risk of cardiovascular disease in postmenopausal breast cancer survivors: two cohort studies in the UK and US

Objective Examine the effect of tamoxifen and aromatase inhibitors on 12 clinically relevant individual cardiovascular outcomes in postmenopausal female breast cancer survivors using large-scale datasets from the UK and US. Design Two prospective cohort studies Setting Population-based using data from the UK Clinical Practice Datalink linked with Hospital Episode Statistics (2002-2016), and the US Surveillance, Epidemiology and End Results-Medicare database (2008-2013). Participants 10005 and 22027 postmenopausal women with breast cancer in the UK and US respectively. Exposures Aromatase inhibitor compared with tamoxifen use; the US cohort additionally included a comparison with an “unexposed” group of women with oestrogen or progesterone receptor positive breast cancer but no endocrine therapy use. Outcomes 12 clinically relevant individual cardiovascular outcomes (and two composite coronary and venous thromboembolic outcomes) Results In both the UK and the US, there was evidence of an increased risk of coronary artery disease in aromatase inhibitor compared with tamoxifen users (UK incidence rate: 10.18 vs 6.87 per 1000 person-years, HR: 1.29, 0.94-1.76; US incidence rate: 35.26 vs 26.95 per 1000 person-years, HR: 1.29, 1.06-1.55), but the US data showed no increase in risk compared with the unexposed group (incidence rate for tamoxifen vs unexposed: 26.95 vs 38.70 per 1000 person-years, HR: 0.74, 0.60-0.92; incidence rate for aromatase inhibitors vs unexposed: 35.26 vs 28.70, HR: 0.96, 0.83-1.10). Similar patterns were seen for other cardiovascular outcomes such as arrhythmia, heart failure, and valvular heart disease. As expected, there were more venous thromboembolic events in tamoxifen users compared with both aromatase inhibitor users and those unexposed. There was a high degree of consistency between results in the two countries. Conclusions Increased risks of several cardiovascular diseases among aromatase inhibitor compared with tamoxifen users appeared to be driven by protective effects of tamoxifen, rather than toxic effects of aromatase inhibitors. We also confirmed the known increased risk of venous thromboembolic events in tamoxifen users. WHAT THIS PAPER ADDS What is already known on this topic It is known that tamoxifen use increases venous thromboembolism risk, but evidence for other cardiovascular outcomes is less clear. Patterns of results are suggestive of a lower risk of coronary heart disease outcomes with tamoxifen compared to both aromatase inhibitor use and no tamoxifen or placebo, but cardiovascular events are often a secondary consideration and inconsistently reported in trials, and most observational studies use composite cardiovascular definitions, ignoring potentially differential effects on specific cardiovascular outcomes. What this study adds Among postmenopausal women with breast cancer, we found an increased risk of several cardiovascular diseases in aromatase inhibitor compared with tamoxifen users across two countries, which appeared to be driven by protective effects of tamoxifen, rather than toxic effects of aromatase inhibitors. We also found the known increased venous thromboembolism risk in tamoxifen users. There was no evidence that aromatase inhibitors or tamoxifen increases cardiovascular disease risk, other than the known increased venous thromboembolism risk with tamoxifen use. However, there was an apparent protective effect of tamoxifen on other cardiovascular outcomes.