Attenuated peripheral endothelial function among women treated with aromatase inhibitors for breast cancer.

Abstract

Data on long-term cardiovascular effects of aromatase inhibitors (AIs) are limited and conflicting. We sought to evaluate the effect of AIs on peripheral endothelial function in patients with breast cancer. This is an observational, prospective study of postmenopausal women with breast cancer who were enrolled at the initiation of cancer treatment. All participants underwent baseline and 6-12 months of follow-up, with peripheral endothelial function testing to measure reactive hyperemia index (RHI). The primary aim was to assess endothelial function deterioration between baseline and follow-up. The secondary aim was to assess the correlation of cardiovascular risk factors with RHI change in women treated with versus without AIs. Among 97 patients, mean (SD) age was 66 (7) years; 59 (61%) women had AI treatment, and 38 women did not (control group). There were no significant differences in baseline characteristics between the groups. Mean (SD) RHI at baseline in the treatment group did not differ significantly from that in the control group [2.2 (0.6) vs. 2.1 (0.5); P=0.15]. The mean (SD) time between baseline and follow-up studies was 262 (60) days. RHI deterioration, evaluated as a dichotomous variable with a 20% cutoff, was significantly more common in the AI group [17 (29%) vs. 4 (11%); P=0.04]. After adjusting for age, treatment with AIs was significantly associated with an RHI deterioration of at least 20% from baseline (odds ratio: 3.6; 95% confidence interval: 1.10-12.07; P=0.03). Further, in the intervention group, women with at least three traditional cardiovascular risk factors were more likely to have RHI deterioration compared to women with λ2 risk factors [10 (42%) vs. 7 (20%); P=0.04]. Amongst women with three or more cardiovascular risk factors, the percentage with RHI deterioration was higher in the AI group than the control group [10/24 (42%) vs. 3/22 (14%); P=0.04], whereas in women with up to two risk factors, the percentage with RHI deterioration was similar between the groups [7/35 (20%) vs. 1/16 (6%); P=0.21]. This study suggests that AIs may be associated with vascular injury. The effect is more pronounced among women with a higher baseline cardiovascular risk factor burden. These findings have potentially important implications, particularly among women at high risk for cardiovascular disease who are treated with AIs for hormone receptor-positive breast cancer.