Centre for Clinical Epidemiology, Lady Davis Institute, Jewish General Hospital, Montreal, Quebec, Canada (F.K.K., K.B.F., S.S., L.A.).
Department of Epidemiology, Biostatistics, and Occupational Health (F.K.K., K.B.F., S.S., L.A.), McGill University, Montreal, Quebec, Canada.
Circulation. 2020 Feb 18;141(7):549-559. doi: 10.1161/CIRCULATIONAHA.119.044750. Epub 2020 Feb 17.
The association between aromatase inhibitors and cardiovascular outcomes among women with breast cancer is controversial. Given the discrepant findings from randomized controlled trials and observational studies, additional studies are needed to address this safety concern.
We conducted a population-based cohort study using the UK Clinical Practice Research Datalink linked to the Hospital Episode Statistics and Office for National Statistics databases. The study population consisted of women newly diagnosed with breast cancer initiating hormonal therapy with aromatase inhibitors or tamoxifen between April 1, 1998, and February 29, 2016. We usedCox proportional hazards models with inverse probability of treatment and censoring weighting to estimate hazard ratios (HRs) with 95% CIs comparing new users of aromatase inhibitors with new users of tamoxifen for each of the study outcomes (myocardial infarction, ischemic stroke, heart failure, and cardiovascular mortality).
The study population consisted of 23 525 patients newly diagnosed with breast cancer, of whom 17 922 initiated treatment with either an aromatase inhibitor or tamoxifen (8139 and 9783, respectively). The use of aromatase inhibitors was associated with a significantly increased risk of heart failure (incidence rate, 5.4 versus 1.8 per 1000 person-years; HR, 1.86 [95% CI, 1.14-3.03]) and cardiovascular mortality (incidence rate, 9.5 versus 4.7 per 1000 person-years; HR, 1.50 [95% CI, 1.11-2.04]) compared with the use of tamoxifen. Aromatase inhibitors were associated with elevated HRs, but with CIs including the null value, for myocardial infarction (incidence rate, 3.9 versus 1.8 per 1000 person-years; HR, 1.37 [95% CI, 0.88-2.13]) and ischemic stroke (incidence rate, 5.6 versus 3.2 per 1000 person-years; HR, 1.19 [95% CI, 0.82-1.72]).
In this population-based study, aromatase inhibitors were associated with increased risks of heart failure and cardiovascular mortality compared with tamoxifen. There were also trends toward increased risks, although nonsignificant, of myocardial infarction and ischemic stroke. The increased risk of cardiovascular events associated with aromatase inhibitors should be balanced with their favorable clinical benefits compared with tamoxifen.
芳香酶抑制剂与乳腺癌女性心血管结局之间的关联存在争议。鉴于随机对照试验和观察性研究得出的结果不一致，需要进一步的研究来解决这一安全性问题。
我们使用英国临床实践研究数据链接到医院事件统计和国家统计局数据库进行了一项基于人群的队列研究。研究人群包括 1998 年 4 月 1 日至 2016 年 2 月 29 日期间新诊断为乳腺癌并开始接受芳香酶抑制剂或他莫昔芬激素治疗的女性。我们使用逆概率治疗和删失加权的 Cox 比例风险模型来估计新使用者与新使用者的风险比（HR）（95%置信区间），比较每个研究结果（心肌梗死、缺血性中风、心力衰竭和心血管死亡率）中芳香酶抑制剂和他莫昔芬的使用情况。
该研究人群包括 23525 名新诊断为乳腺癌的患者，其中 17922 名患者开始接受芳香酶抑制剂或他莫昔芬治疗（分别为 8139 名和 9783 名）。与使用他莫昔芬相比，使用芳香酶抑制剂与心力衰竭（发生率为 5.4 比 1.8 每 1000 人年；HR，1.86 [95%CI，1.14-3.03]）和心血管死亡率（发生率为 9.5 比 4.7 每 1000 人年；HR，1.50 [95%CI，1.11-2.04]）风险显著增加相关。与使用他莫昔芬相比，芳香酶抑制剂与心肌梗死（发生率为 3.9 比 1.8 每 1000 人年；HR，1.37 [95%CI，0.88-2.13]）和缺血性中风（发生率为 5.6 比 3.2 每 1000 人年；HR，1.19 [95%CI，0.82-1.72]）的 HR 升高，但置信区间包含零值。
在这项基于人群的研究中，与他莫昔芬相比，芳香酶抑制剂与心力衰竭和心血管死亡率风险增加相关。尽管无统计学意义，但心肌梗死和缺血性中风的风险也呈上升趋势。与芳香酶抑制剂相关的心血管事件风险增加应与其与他莫昔芬相比的有利临床获益相平衡。
